Skip to main content

Full text of "Bulletin of the British Ornithologists' Club"

See other formats


Bulletin of the 



British 




ik 



j^ 



Ornithologists' 



19 JUN 2006 




Volume 126 Supplement 2006 
Recent Avian Extinctions 

Edited by Guy M. Kirwan 



Bull. B.O.C. 2006 126A 



Bulletin of the 

BRITISH ORNITHOLOGISTS' CLUB 

Vol. 126A Published 5 June 2006 



Recent Avian Extinctions 



Papers from a conference of this title 

organised by the British Ornithologists' Union, 

and supported by the British Ornithologists' Club 

and Linnaean Society of London, 

held at the Linnean Society of London offices, 

on 1 November 2004 



Edited by Guy M. Kirwan 



TKENATURAL 

KigTOf 'ft MW 

19 JUN 2006 

PRESENTED 




frontispiece. Recently extinct passerines from the Hawaiian island of Oahu, from top to bottom: Akepa 
Loxops cot (incus rufa (extinct 1893), Oahu Thrush Myadestes oahensis (extinct 1825), Nukupu'u 
Hemignathus lucidus lucidus (extinct c.1860), Ou Psittirostra psittacea (extinct c.1900) Akialoa 
Hemignathus lichtensteini (extinct 1837) and Oo Moho apicalis (extinct 1837) (Julian Pender Hume) 



3 Bull. B.O.C. 2006 126A 

Recent Avian Extinctions 

by Julian Hume 

The idea for a one-day conference on recent avian extinctions in conjunction with 
the Linnaean Society of London was originally proposed by Chris Feare. The 
conference was to comprise a presentation of related topics from researchers 
working on predominantly Holocene extinctions, particularly as we enter an era of 
increasingly catastrophic worldwide extinction events. Chris unfortunately had to 
withdraw at an early stage and I was presented the opportunity to guide the project 
through to completion. My aim therefore was to organise an international one-day 
symposium that reflected the present diverse and multi-disciplinary approach to 
avian research including natural extinction events, DNA, statistical analysis and 
evidence derived from morphological and historical studies. 

In planning, the conference was mainly constructed to encourage discussion and 
possible scientific collaboration. The choice of papers was biased towards oceanic 
islands, as evidence derived from the palaeontological record indicates that 
hundreds, if not thousands, of island bird species have disappeared in recent historic 
times, nearly all as a result of anthropogenic activity (e.g. Steadman 1995). 
Therefore, a conference dealing exclusively with recent avian extinctions provided 
an ideal opportunity to present an overview of research ranging from case studies of 
species to methods of identifying the scope of extinction events. As a result, I hope 
that these proceedings highlight the importance of obtaining data through a number 
of disciplines, and emphasise the fact that biogeographic conclusions based on 
current diversity and distribution of both oceanic island and, increasingly, 
continental birds is very much an artefact of human interference. 

This conference would not have taken place had it not been for the commitment 
of Steve Dudley, of the British Ornithologists' Union (BOU), the British 
Ornithologists' Club (BOC) and the Linnaean Society of London. In particular the 
BOU provided logistical and administrative expertise and the Linnaean Society their 
support and premises for the occasion. I am also indebted to the BOC, who funded 
both this supplement and the colour artwork included within. I further thank Guy 
Kirwan and Robert Prys-Jones for their assistance during the composition of this 
publication, and all of the speakers/attendees who made the day such a memorable 
one. Finally, I take this opportunity to dedicate these proceedings to the late Prof. 
Janet Kear, whose passion and commitment to avian research was unprecedented. 
Janet co-authored a paper for the conference but sadly, due to ill health, could not 
personally attend the meeting, and passed away a short time after. 

Reference: 

Steadman, D.W. 1995. Prehistoric extinctions of Pacific island birds: biodiversity meets zooarchaeology. 
Science 267: 1123-1131. 



Michael Bunce & Richard Y. Holdaway 4 Bull. B.O.C. 2006 126A 

New Zealand's extinct giant eagle 

by Michael Bunce & Richard N. Holdaway 

Since Richard Owen described the first moa (Dinornithidae) bone in 1839 the 
extinct avifauna of New Zealand has continued to offer biological 'surprises' — the 
Haasfs Eagle Harpagornis moorei is no exception. A fact about New Zealand still 
not widely appreciated is that before human occupation it had no native mammals 
(other than three species of bat) — therefore its flora and fauna evolved in ways very 
different to those seen in other ecosystems. In terms of a place to research recent 
avian extinctions, New Zealand offers a good case study— 50% of the nearly 250 
bird species (present 1 ,000 years ago) are now extinct. The bones of these extinct 
taxa can be found in caves and swamp deposits throughout New Zealand — and this 
excellent avian fossil record extends well into the last glacial period (30,000 years 
bp. The relatively cool climate has meant that the bones, and the DNA in them, are 
well preserved. Advances in ancient DNA techniques in the last decade have 
provided new avenues of research that enable us to investigate the genetic legacy of 
extinct birds like the Haast's Eagle. 

Bones of the Haast's Eagle were first discovered by Julius Haast (founder and 
first director of the Canterbury Museum) during excavations of the Glenmark 
swamp in 1871. Haast recognised immediately that this was a huge raptorial bird 
(modern estimates are 10-14 kg and a wingspan of up to 3 m) that exceeded the size 
of any extant eagle by some margin. Relatively little research was conducted on the 
eagle in the ensuing 100 years, but its sheer size fuelled speculation regarding the 
species' life history, whether it flew and its diet. In the 1990s it was realised that 
punctures in the pelves of moa collected years before perfectly matched the size and 
spacing of the Haast's Eagle talons (Fig. la, p. 6). Moa up to 200 kg in weight were 
targets for this predator — the bones show that the eagle struck from the side with 
considerable force, gripped the moa's pelvic area with one foot, and killed with a 
single strike by the other foot to the neck or head (Fig. 1). The eagle's extinction 
(c.500 years ago) was closely tied to the demise of the moa. Rock art, Maori oral 
history and bone artefacts prove early Polynesians coexisted with the eagle, but 
there is no evidence that humans were targets for this aerial predator. 

Much of New Zealand's avifauna has relatives in Australia — that of Haast's 
Eagle was long suspected to be the large (4.5 kg, 2 m wingspan) Wedge-tailed Eagle 
Aquila audax — which seemed a good analog because it had been known to attack 
young emus Apteryx — large flightless birds that are distantly related to moas. An 
exploratory skeletal analysis, using representative genera within the Accipitridae 
(but lacking any Australasian representative of the genera Hieraaetus), placed 
Haast's Eagle as sister to the Wedge-tailed Eagle. However, shifts in body size are 
common in island ecosystems and may distort skeletal characters used in 
phylogenetic reconstructions. 



Michael Bunce & Richard N. Holdaway 5 Bull. B.O.C. 2006 126 A 

To investigate the genetic history of this fascinating raptor we sampled two eagle 
bones (c. 3,000 years old) from the collections of the Museum of New Zealand Te 
Papa Tongawera. With only 70 or so individuals ever found and only two relatively 
complete skeletons in existence these are rare bones, making us grateful to the 
museum for permitting us to destructively sample their collection. The preliminary 
gene sequences were so different from our original expectations that we initially 
questioned their authenticity. These data indicated that Wedge-tailed Eagle was only 
distantly related to Haast's Eagle, which was in fact related to some of the world's 
smallest eagles — the Little Eagle Hieraaetus morphnoides from Australia and New 
Guinea, and the Eurasian Booted Eagle H. pennatus, which typically weigh less 
than 1 kg. Even more striking was how closely genetically related the eagle species 
were (1.25% difference in their mitochondrial cytochrome-^ gene). In accordance 
with a conservative molecular clock, we estimate that their common ancestor lived 
c. 1 MYA. These data suggests an eagle arrived in New Zealand and increased in 
weight by 10-15 times during this period. This spectacular evolutionary change 
illustrates the potential speed of size alteration within lineages of vertebrates, 
especially in island ecosystems. 

The question remains as to what environmental/ecological factors drove the 
eagle to the upper limits for powered (flapping) flight from an ancestor that was 
likely one-tenth its size? Answering questions concerning the life history of extinct 
megafauna is akin to guesswork, but we speculate that size of available prey and the 
absence of other large predators were key in driving the size increase — after killing 
a moa, the eagle could have fed unhindered and perhaps remained at a kill site for 
days. 

The DNA phylogeny we constructed from 1 6 extant eagle species demonstrated 
considerable problems with the current classification of 'booted' eagles (those with 
feathered tarsi), especially within the genera Aquila, Hieraaetus and Spizaetus 
which are clearly paraphyletic. Assignment of species limits within these genera has 
traditionally been problematic, making our observation expected. However, it is 
apparent the name for the extinct New Zealand Eagle should be amended to 
Hieraaetus moorei. 

In summary, ancient DNA represents a valuable means by which to study the 
genetic history of extinct fauna, but is only one piece of the puzzle in unravelling 
their life history. An integrated approach combining DNA, stable isotopes, 
morphological analysis and other technologies represents the only way to 
investigate the multitude of other questions concerning extinct fauna, e.g. how large 
was the population of Haast's Eagle? 

A fully referenced research paper is available for free download from the PLos 
biology website, www.plosbiology.org. 

Addresses: Michael Bunce, Department of Anthropology, McMaster University, ON, L8S 4L9, Canada, 
e-mail: buncem@mcmaster.ca. Richard N. Holdaway, Palaecol Research Ltd, RO. Box 16569, 
Christchurch, New Zealand, e-mail: piopio@paradise.net.nz 



Michael Buncc & Richard N. Holdaw< 



Bull. B.O.C. 2006 126A 




^ 



Figure 1 . New Zealand's extinct Haast's Eagle Hieraaetus moorei (formerly Harpagornis moorei). (A) 
moa pelvis showing puncture marks (marked by arrows) caused by an eagle strike. Evidence of eagle 
strikes are preserved on skeletons of moa (Dinornithidae) weighing up to 200 kg (Trevor Worthy). (B) A 
direct comparison of the claws of Haast's Eagle with those of its closest known relative, Little Eagle 
Hieraaetus morphnoides, demonstrating the extent of the size change. (C) Artist's impression of Haast's 
Eagle attacking the extinct New Zealand moa. The eagle struck and gripped the moa's pelvic area (A), 
and then killed with a single strike of the other foot to the head or neck (John Megahan) 



S. H. M. Butchart et al. 7 Bull. B.O.C. 2006 126 A 

Going or gone: defining 'Possibly Extinct 5 species 
to give a truer picture of recent extinctions 

by 5. H. M. Butchart, A. J. Stattersfield &T. M. Brooks 

The IUCN Red List is widely regarded as the most authoritative classification of 
species by their extinction risk (Lamoreux et al. 2003, Hambler 2004, Rodrigues et 
al. 2006), including those species known to have become extinct in recent times. 
Birds are the best-documented class of organisms on the Red List, and the fourth 
complete assessment of the status of the world's birds was recently published 
(BirdLife International 2004, IUCN 2004), and updated (at www.birdlife.org) for 
the 2005 IUCN Red List. As well as 1,208 threatened bird species in the categories 
of Critically Endangered, Endangered and Vulnerable (in order of decreasing risk of 
extinction), it lists 131 species as having become Extinct since 1500 (for which 
'there is no reasonable doubt that the last individual has died': IUCN 2001), and an 
additional four species as Extinct in the Wild ('known only to survive in captivity': 
IUCN 2001). 

However, extinction — the disappearance of the last individual of a species — is 
very difficult to detect (Diamond 1987). For a species to be listed as Extinct requires 
that exhaustive surveys have been undertaken in all known or likely habitat 
throughout its historic range, at appropriate times (diurnal, seasonal, annual) and 
over a timeframe appropriate to its life cycle and life form (IUCN 2001). Listing as 
Extinct has significant conservation implications, because conservation funding is, 
justifiably, not targeted at species believed extinct. Following a precautionary 
approach, conservationists are therefore reluctant to designate species as Extinct if 
there is any reasonable possibility that they may still be extant, in order to avoid the 
'Romeo Error' (Collar 1998), where we might give up on a species before it is too 
late. This term was first applied to the case of Cebu Flowerpecker Dicaeum 
quadricolor, which was rediscovered in 1992 after 86 years without a record 
(Dutson et al. 1993), having been written off as extinct at least 40 years earlier on 
the presumption that no forest remained on the island of Cebu (Magsalay et al. 
1995). This remarkable rediscovery is by no means unique. For example, Jerdon's 
Courser Rhinoptilus bitorquatus was rediscovered in 1986 also after 86 years 
without a record (Bhushan 1986). Caerulean Paradise-flycatcher Eutrichomyias 
rowleyi was known only from the 1878 type specimen and a belatedly published 
sight record in 1978, with fruitless searches in 1985-86 (Whitten et al. 1987) prior 
to its rediscovery in 1998 (Riley & Wardill 2001). 

On the other hand, for some Critically Endangered species the chances of 
rediscovering a population must be extremely low, and in all probability they are 
already extinct. For example, Alaotra Grebe Tachybaptus rufolavatus underwent a 
well-documented decline owing to incidental mortality in monofilament gill-nets 
and predation by introduced carnivorous fish, compounded by hybridisation with 
Little Grebe T. ruficollis. The last confirmed records were in 1985, with individuals 



V // \/. Butchart et al. 8 Bull. B.O.C. 2006 126A 

showing some characters of the species seen in 1986 and 1988 (Hawkins et al. 
2000). The species was near-flightless and restricted to the Lake Alaotra area. There 
is a slim chance that individuals could survive at Lake Amparihinandriambavy, 
where unidentified grebes were seen in 2000, but this species is in all probability 
now extinct (BirdLife International 2004). Similarly, Nukupu'u Hemignathus 
lucidus is endemic to the Hawaiian Islands where it has not been recorded since 
1 995-96 despite extensive effort in a large proportion of the historic range (Pratt et 
al. 2001). It is in all likelihood extinct as a result of habitat loss and degradation 
combined with introduced diseases such as avian malaria spread by introduced 
mosquitoes. 

A precautionary approach by IUCN to classifying extinctions is appropriate in 
order to encourage continuing conservation efforts until there is no reasonable doubt 
that the last individual of a species has died. It also minimises the danger of 'crying 
wolf and reducing confidence in the accuracy of the label Extinct. However, this 
approach biases analyses of recent extinctions based only on those species officially 
classified Extinct or Extinct in the Wild. For example, the number of recent 
extinctions documented on the IUCN Red List is likely to be a significant 
underestimate, even for well-known taxa such as birds. In recognition of this, we 
develop a framework to examine relevant evidence and judge as objectively as 
possible which Critically Endangered species are likely to be already extinct. Using 
data on these species and on species evaluated as Extinct and Extinct in the Wild, 
we re-analyse recent extinctions to provide a more realistic assessment of their rate, 
taxonomic distribution, geography and causes. 

Methods 

Information on Extinct, Extinct in the Wild, and Critically Endangered species were 
taken from BirdLife International (2004), updated at www.birdlife.org. The 
accounts for Extinct species in BirdLife International (2004) were based largely on 
those in Brooks (2000). Dates were assigned to extinctions and possible extinctions 
based on the date of the last reliable or confirmed record. In cases for which 
extinction was estimated to have occurred during a particular period, the midpoint 
was taken. In theory, more sophisticated techniques for estimating extinction dates 
are available (Solow 1993), but these require knowledge of the dates of multiple 
records of a species prior to its extinction, which are rarely available for extinct 
birds. Recognising that it is difficult in most cases to precisely date extinctions, we 
analysed temporal patterns by pooling data into 25- or 50-year intervals. We 
analysed the taxonomy of recent extinctions at the family level, using binomial one- 
tailed tests to compare the significance of differences between the percentages of 
extinct species per family with the percentage for the class Aves. Causes of 
extinction and threats to extant threatened species were coded according to a 
standard classification of threats used to document all threatened species on the 
IUCN Red List (http://www.redlist.org/info/major_threats.html). For the purposes 
of the analyses here, threats deriving from alien invasive species impacting the 



S. H. M. Butchart et al. 9 Bull. B.O.C. 2006 126A 

habitat of a threatened or extinct species were pooled with other forms of threat by 
invasive species, rather than with other forms of habitat degradation. For the 
comparison of extinct and extant threatened species, we considered for the latter 
only high and medium-impact threats, i.e. those that affect the majority of the 
population and cause rapid declines (BirdLife International 2004). 

Defining 'Possibly Extinct' species 

We defined 'Possibly Extinct' species as those that are, on the balance of evidence, 
likely to be extinct, but for which there is a small chance that they may be extant 
and thus should not be listed as Extinct until adequate surveys have failed to find 
the species and local or unconfirmed reports have been discounted. 'Possibly 
Extinct in the Wild' correspondingly applies to such species known to survive in 
captivity. 

For each species we considered five main types of evidence for extinction: 

• For species with recent last records, the decline has been well documented. 

• Severe threatening processes are known to have occurred (e.g. extensive habitat 
loss, the spread of alien invasive predators, intensive hunting, etc.). 

• The species possesses attributes known to predispose taxa to extinction, e.g. 
natural rarity and/or tiny range (as evidenced by paucity of specimens relative to 
collecting effort), flightlessness, allospecies or congeners that may have become 
extinct through similar threatening processes, etc. 

• Recent surveys have been apparently adequate given the species' ease of 
detection, but have failed to detect the species. 

We considered four types of evidence against extinction: 

• Recent field work has been inadequate (any surveys have been insufficiently 
intensive/extensive, or inappropriately timed; or the species' range is 
inaccessible, remote, unsafe or inadequately known). 

• The species is difficult to detect (it is cryptic, inconspicuous, nocturnal, 
nomadic, silent or its vocalisations are unknown, identification is difficult, or the 
species occurs at low densities). 

• There have been reasonably convincing recent local reports or unconfirmed 
sightings. 

• Suitable habitat (free of introduced predators and pathogens if relevant) remains 
within the species' known range, and/or allospecies or congeners may survive 
despite similar threatening processes. 

By explicitly laying out and classifying evidence for and against extinction 
under this framework, we then judged where to place each species on a continuum 
from high to low confidence of extinction, on a spectrum from Extinct to Critically 
Endangered (Possibly Extinct) to Critically Endangered. For any given balance of 
evidence, the position on this continuum was influenced by the time since the last 



5. H. M. Buichart el al. 



10 



Bull. B.O.C. 2006 126 A 



continued record (see Fig. 1). For example, for species with recently confirmed 
records to be placed at the Extinct end of the spectrum, there had to be greater 
confidence in the extinction, i.e. greater confidence in the adequacy of surveys, the 
absence or inadequacy of local/unconfirmed records, greater severity of threatening 
processes, and better documentation of, and confidence in, observed population 
declines. In contrast, species that had not been recorded for many decades (e.g. more 
than 100 years) were judged to be more likely to have become extinct for a given 
balance of evidence for and against extinction, owing to the sheer length of time 
without records. Deciding the strength of evidence for and against extinction is 
necessarily subjective. However, this framework helped to make these judgements 
as objective as possible, by setting out the evidence, and weighing this against the 
time since the last confirmed record. 

We tested this framework on 40 Critically Endangered bird species that we 
considered candidates for Possibly Extinct status. This included all species for 
which there was any reasonable possibility that they might be extinct, including any 
that had not been seen for >10 years (despite reasonable searches and/or for which 
there was a plausible threatening process), and any that had been last seen <10 years 



High 



Low 



Low 



High 



Atitlan Grebe 
Aldabra Warbler 



Night Parrot 




Critically 
Endangered 



White-chested White-eye 
Pohnpei Mountain Starling 



Magdalena 
Tinamou 



Himalayan Quail 
Samoan Moorhen 



c.10 



.50 



>100 



No. of years since last record 



Figure 1 . Schematic showing, with selected examples, how time since last record interacts with 
confidence of extinction to determine how species are classified as Critically Endangered, Possibly 
Extinct or Extinct. For species last recorded quite recently there needs to be greater confidence that the 
hist individual has died in order for the species to be placed at the extinct end of the spectrum from 
Critically Endangered to Extinct. 



S. H. M. Butchart et al. 1 1 Bull B.O.C. 2006 126A 

ago for which there had been a well-documented decline of a tiny population. Of 
these, we identified 15 as Possibly Extinct (including one Possibly Extinct in the 
Wild species; Appendix 1) and 25 as Critically Endangered (Appendix 2). 

One-third of the Possibly Extinct species have not been recorded for more than 
50 years or so, and this significant duration since the last records is, of itself, strong 
evidence that these species may well be extinct. For example, Hooded Seedeater 
Sporophila melanops is known only from the type specimen collected over 180 
years ago (BirdLife International 2004). Although habitat destruction in the region 
of the type locality has not been exceptionally severe, the sheer duration of time 
without records of a species that could be expected to be relatively easily identified 
and detected can be considered strong evidence that the species is now extinct. 
Similarly, Guadalupe Storm-petrel Oceanodroma macrodactyla has not been 
recorded since 1912 despite several searches, following a severe decline owing to 
predation by introduced cats and habitat degradation by introduced goats (BirdLife 
International 2004). Only the difficulty of detecting storm-petrels at their breeding 
colonies at night (when the birds are active) and the continued survival of other 
storm-petrels on the island point to the possibility that some individuals survive (and 
hence that classification as Extinct would be premature). 

The remaining Possibly Extinct species have undergone well-documented 
declines, with the most recent records in the last 25 years or so. For example, the 
last known Spix's Macaws Cyanopsitta spixii were monitored until the last 
individual disappeared in 2000, following a severe decline owing to unsustainable 
and intensive exploitation for the cagebird trade (Juniper 2003). Searches have not 
led to the discovery of any other populations, although it is conceivable, if unlikely, 
that further individuals survive. Similarly, the last well-documented sighting of 
Oloma'o Myadestes lanaiensis was in 1980, with an unconfirmed report in 1988, 
and there have been no subsequent records despite further surveys in most of the 
historical range. It is likely to have been driven extinct by disease spread by 
introduced mosquitoes, and as a result of habitat destruction (Reynolds & 
Snetsinger 2001). However, the remote Oloku'i Plateau has not been surveyed 
recently and could still harbour some birds. 

Three Vulnerable species have not been recorded for many years, but in each 
case the threats to them are less intense, and the lack of records clearly results from 
a lack of surveys, taxonomic uncertainties and/or identification difficulties, rather 
than because of possible extinction. They are classified as Vulnerable rather than 
Critically Endangered owing to their presumed small (rather than tiny) and 
declining populations. The species are: Nicobar Sparrowhawk A ccipiter butleri (last 
definite record 1901; possible sightings in the 1990s, but identification uncertain 
owing to confusion with Besra A. virgatus); Manipur Bush-quail Perdicula 
manipurensis (last definite record 1932; possible record in 2004, and cessation of 
hunting, lack of field work and difficulty of detecting this species are likely to 
explain the lack of records); and Black-browed Babbler Malacocincla perspicillata 
(known only from a specimen collected in 1843-48, but the lack of subsequent 



S. //. \I. Butchart el al. 12 5////. 5.O.C. 2006 126A 

records is most likely to have been a result of confusion over its taxonomic status). 
In addition, three Endangered species have also not been recorded recently, but are 
regarded as likely to be extant for similar reasons. They are classified as Endangered 
on the basis of their small known ranges and because their remaining populations 
are assumed to be too large to qualify as Critically Endangered. They are: Recurve- 
billed Bushbird Clytoctantes alixii (last recorded 1965 despite recent searches, but 
known from several sites in north Colombia and north-west Venezuela), Chestnut- 
bellied Flowerpiercer Diglossa gloriosissima (last recorded in 1965, but there has 
been a dearth of recent field work within its known range in Colombia), and Tachira 
Antpitta Grallaha chthonia (last recorded 1956 despite recent searches, but suitable 
habitat remains within the large national park in Venezuela from which the species 
is known). 

We also examined a number of Data Deficient species that have not been 
recorded for many years. Data Deficient is a category on the IUCN Red List applied 
to species for which 'there is inadequate information to make a direct or indirect 
assessment of [the] risk of extinction' (IUCN 2001). For six species (Cayenne 
Nightjar Caprimulgus maculosus, Vaurie's Nightjar C. centralasicus, White-chested 
Tinkerbird Pogoniulus makawai, Red Sea Swallow Hirundo perdita, Sillem's 
Mountain-finch Leucosticte sillemi and Black-lored Waxbill Estrilda nigriloris) the 
available evidence suggests that they are unlikely to be threatened (and hence 
unlikely to be near extinction or potentially extinct), because no threatening factor 
is known or can be inferred, and there are convincing practical reasons for the lack 
of recent records (e.g. surveys have been inadequate, the species is difficult to detect 
and/or there is taxonomic uncertainty). In three cases (Sharpe's Rail Gallirallus 
sharpei, Coppery Thorntail Popelairia letitiae and Bogota Sunangel Heliangelus 
regalis) knowledge of the original range is so poor that no further inferences can be 
made (e.g. Sharpe's Rail is known from an 1893 specimen of unknown provenance, 
possibly from the Greater Sundas). 

The 1 5 species we identified as Possibly Extinct will be tagged as such on the 
IUCN Red List. The framework developed here is currently being tested on 
amphibians and mammals, prior to being considered, with potential modifications, 
for general adoption by the IUCN Red List. 

Recent extinctions reanalysed 

We combined data on Critically Endangered (Possibly Extinct), Extinct and Extinct 
in the Wild species from BirdLife International (2004; updated at www.birdlife.org) 
to undertake a realistic analysis of the pattern of recent extinctions. 

Extinction rates 

Combining totals for Extinct (131), Extinct in the Wild (four) and Critically 
Endangered (Possibly Extinct) species (15), exactly 150 bird species have gone or 
arc likely to have become extinct since 1500. This represents a rate of 0.30 species 
per year. Since 1900, the total is 59 species: 0.56 species per year. While these data 



S. H. M. Butchart et al. 



13 



Bull. B.O.C. 2006 126 A 



20 

E 18 

S 16 

£ 14 

2 12 

Q- 10 



■ , . .mill 




n 

II 



nPE 

■ EX/EW 



Figure 2. Number of avian extinctions per 25-year period showing totals for Critically Endangered 
(Possibly Extinct) species ('PE'; «=15), and Extinct ('EX'; «=131) plus Extinct in the Wild ('EW; «=4) 
species. 



may underestimate the extinction rate of 500 years ago, because some species may 
have become extinct without our knowledge (Balmford 1996), it appears that the 
extinction rate increased rapidly from the late 1600s, and peaked in the late 1800s 
and early 1900s at 0.72 species p. a. (in 1875-1925; Fig. 2). Very recent extinction 
rates remain high: 17 species were lost in the last quarter of the 20th century, and 
two species since 2000. The last known individual of Spix's Macaw Cyanopsitta 
spixii (Critically Endangered [Possibly Extinct in the Wild]) disappeared in Brazil 
in late 2000, and the last two known individuals of Hawaiian Crow Corvus 
hawaiiensis (Extinct in the Wild) disappeared in June 2002. Po'o-uli Melamprosops 
phaeosoma, also from the Hawaiian Islands, looks set to become the next addition 
to this list: one of the last three known individuals was taken into captivity in 
September 2004 but died two months later, and the other two individuals have not 
been seen for over a year (K. Swynnerton in lift. 2004). Fig. 2 shows clearly how 
important it is to consider Possibly Extinct species in assessing recent extinction 
rates: the total number of estimated extinctions in the last quarter of the 20th century 
almost doubles from nine to 1 7 when Possibly Extinct species are included. 

How do these extinction rates compare to those derived from the fossil record? 
Comparisons of absolute rates are difficult given considerable uncertainty over the 
total number of species on the planet, so it is useful to compare relative extinction 
rates, expressed as extinctions per million species per year (E/MSY; Pimm et al. 
1995). Mean fossil species lifetimes produce a background extinction rate of 0.1-1 
E/MSY. The total number of bird extinctions since 1500 (150/9,906 species) 
therefore equates to 30-300 times the background rate. Taking the number of 
extinctions since 1900 (59/9,815 extant species in 1900) gives an extinction rate 
57-570 times background extinction rates. These are still highly conservative 
estimates for the extinction rate across all taxa, because many taxonomic groups 
(e.g. amphibians, fish, plants, invertebrates) have on average much smaller ranges, 
and hence likely higher extinction rates in the face of human impact than do birds. 



S. H. M. Butcharl el al. 



14 



Bull. B.O.C. 2006 126A 



Estimates of extinction rates derived from measurement of a range of extinction 
drivers (e.g. habitat destruction, human energy consumption) yield E/MSY 
1,000-1 1,000 higher than background rates (Pimm & Brooks 1999). 

Geography of recent extinctions 

Recent avian extinctions have occurred across the world, with particularly large 
numbers in Hawaii (27), Mauritius (18), New Zealand (14), Reunion (11) and St 
Helena (nine; Fig. 3). The majority (89.3%) has been on islands even though most 
bird species (>80%) live on continents (Johnson & Stattersfield 1990, Manne et al. 
1999). However, continental species have been far from immune, and those subject 
to extinction often originally had extensive ranges. The wave of extinctions on 
islands may be slowing, perhaps because many of the potential introductions of 
alien species to predator-free islands have already occurred, and because so many 
susceptible island species are already extinct. By contrast, the rate of extinctions on 
continents appears to be sharply increasing (Fig. 4) owing to extensive and 
expanding habitat destruction (see below). 

Taxonomy of recent extinctions 

Recent extinctions have not been random with respect to taxonomy. Thirteen 
families were found to have suffered significantly more extinctions than expected 
by chance (Table 1). Among large families, Anatidae (ducks, geese and swans), 
Rallidae (rails), Psittacidae (parrots) and Sturnidae (starlings) have suffered a 
disproportionate number of extinctions. The Dromaiidae (emus), Raphidae (Dodo 
Raphus cucullatus and solitaires) and Acanthisittidae (New Zealand wrens) have all 
lost 50% or more of their species in the last 500 years. Conversely, some families 




s — ^'y 



Figure 3. Global distribution of recent avian extinctions. Localities show last known records of Extinct 
(squares. /?=131), Extinct in the Wild (circles, n=A), and Critically Endangered (Possibly Extinct) species 
(triangles, // 15). 



S. H. M. Butchart et al. 



15 



Bull. B.O.C. 2006 126 A 



I I i I I I I I Irlrlrlrlllrlrl 



□ Continent 
■ Island 



o o 
m o 

LO CD 



Figure 4. Number of avian extinctions per 25-year period on continents and islands. Totals include 
Extinct («=131), Extinct in the Wild (n=4), and Critically Endangered (Possibly Extinct) species (n=l5). 

(or subfamilies) have suffered significantly fewer extinctions than expected by 
chance: Accipitridae (hawks and eagles, extinctions/ 239 species), Formicariidae 
(antthrushes, 0/267), Furnariidae (ovenbirds, 0/242), Tyrannidae (tyrant-flycatchers, 
0/409), Muscicapidae (thrushes, babblers, warblers and Old World flycatchers, 
12/1,551), Emberizidae (buntings, 1/614; PO.02 in each case). Passerines formed 
19% of continental extinctions (3/16 species) and 34% of island extinctions (46/134 
species), but this difference is not significant (% 2 = 1.58, P=0.21). 



TABLE 1 

Families with significantly more recently extinct species (Extinct, Extinct in the Wild, and Possibly 

Extinct) than expected by chance. 



Family 


No. species 


No. extinct 
species 


% extinct 


P 


Raphidae (dodo, solitaires) 


2 


2 


100 


0.0002 


Dromaiidae (emus) 


3 


2 


66.7 


0.0007 


Acanthisittidae (New Zealand wrens) 


4 


2 


50 


0.0014 


Drepanididae (honeycreepers) 


34 


16 


47.1 


O.0001 


Callaeidae (New Zealand wattlebirds) 


3 


1 


33.3 


0.0450 


Upupidae (hoopoes) 


3 


1 


33.3 


0.0450 


Rallidae (rails) 


156 


23 


14.7 


O.0001 


Podicipedidae (grebes) 


22 


3 


13.6 


0.0044 


Ardeidae (herons) 


67 


4 


6 


0.0193 


Psittacidae (parrots) 


374 


20 


5.3 


O.0001 


Sturnidae (starlings) 


114 


5 


4.4 


0.0308 


Anatidae (ducks, geese, swans) 


164 


7 


4.3 


0.0039 


Columbidae (pigeons) 


318 


13 


4.1 


0.0014 



5. H. M. Bittchan et al. 



16 



Bull. B.O.C. 2006 126A 



Causes of recent extinctions 

Extinction is a natural phenomenon, being the final stage of the evolutionary 
trajectory that each species follows. However, recent extinctions appear to have 
been precipitated by human actions, either directly or indirectly. Here we analyse 
the broad mechanisms by which such extinctions have occurred, as classified on the 
IUCN Red List (BirdLife International 2004, IUCN 2004). 

The impacts of habitat destruction and degradation, alien invasive species and 
over-exploitation by humans have been the major causes of recent avian extinctions 
(Fig. 5). Alien invasive species have been a cause of extinction or likely extinction 
for at least 77 species. Invasive species have impacts in different ways. Most 
important has been predation: introduced dogs, pigs, mongooses and, in particular, 
cats and rats have contributed to the extinction of at least 56 species. The most 
notorious example was the Stephen's Island Wren Traversia lyalli, whose entire 
world population was rapidly wiped out when cats became established on the island 
in 1894 (Tyrberg & Milberg 1991, Galbreath & Brown 2004). Diseases caused by 
introduced pathogens have contributed to the extinction of 20 species, 16 of them on 
Hawaii where introduced avian malaria and avian pox (transmitted by introduced 
mosquitoes) has had (and continues to have) devastating consequences (Scott et al. 
1986, van Riper et al. 1986, Atkinson et al. 1995). Habitat destruction by sheep, 
rabbits and goats has been implicated in the extinctions of another ten species, and 
competitors have impacted six species. Gurevitch & Padilla (2004) argued that the 
evidence for invasive species having contributed to extinctions is poor, and noted that 
just 2% of 762 species listed as Extinct on the 2003 IUCN Red List were documented 
as having been impacted by invasive species. Their result contrasts with ours that 




Cause of extinction 



Figure 5. Causes of recent avian extinctions. Totals include Extinct (n=131), Extinct in the Wild (n=4), 
and Critically Endangered (Possibly Extinct) species (/?=15). 



S. H. M. Butchart et al. 



17 



Bull. B.O.C. 2006 126A 



invasive species were a major contributory factor to 51% of recent avian extinctions. 
Blackburn et al. (2004) and Clavero & Garcia-Berthou (2005) also provided strong 
evidence of the importance of invasive species in driving avian extinctions. 

It is important to note that many species are impacted by combinations of 
threats: 48.7% of extinct species have multiple causes of extinction recorded, and 
this figure is likely to be an underestimate owing to lack of information on historical 
extinctions. 

There are differences in the causes of extinctions of island versus continental 
species, with habitat loss and exploitation appearing to be more important causes of 
extinctions on continents than islands, although this result was marginally non- 
significant (habitat loss: 87.5% vs. 56.0% of species; exploitation: 62.5% vs. 38.1% 
of species; invasive species: 37.5% vs. 53.0% of species; x 2= 4.13, P=0.076; Fig. 
6). The apparent reduced importance of exploitation as an extinction driver on 
islands may be partly explained by the fact that passerines (which, being smaller, are 
less often targets for hunting) form a substantially lower proportion of island 
extinctions compared to continental extinctions (see above). It may also be a 
consequence of an extinction filter effect (Balmford 1996): non-passerine island 
species susceptible to exploitation through their size and naivete may have already 
been driven extinct prior to 1500. 

It is interesting to compare the threats to Extinct and Possibly Extinct species 
with those to extant threatened species (Fig. 7). Whilst habitat loss is the most 
important factor in both cases (impacting 59.3% of extinct species and 54.6% of 
threatened species), invasive species and exploitation were much more important as 




■ Continent 
□ Island 



T> O 

£ E 



Cause of extinction 



Figure 6. Causes of recent avian extinctions on continents («=16 species) and islands («=134 species). 
Totals include Extinct («=131), Extinct in the Wild (w=4), and Critically Endangered (Possibly Extinct) 
species (»=15). 



S. H. M. Butchart et al. 



18 



Bull B.O.C. 2006 126A 




o 30 



■ Extinct species 

□ Other threatened species 




Threat/cause of extinction 

Figure 7. Causes of recent avian extinctions compared to threats to extant threatened birds. Extinct 
species include Extinct («=131), Extinct in the Wild (n=4), and Critically Endangered (Possibly Extinct) 
species (n=\5). Other threatened species include those classified as Critically Endangered (excluding 
Possibly Extinct), Endangered and Vulnerable («=1,193). 

causes of extinctions (implicated for 51.3% and 41.3% of species respectively) than 
as a threat to extant threatened species (12.1% and 13.1% of species respectively). 
However, as Blackburn et al. (2004) pointed out, invasive species (particularly 
predators) are still a potentially important driver for future extinctions. Most islands 
currently have few invasive predators: colonisation by additional predators is likely 
to lead to progressively more extinctions unless prompt intervention is achieved. 



25 



n 20 



15 



£ 10 

E 



■ Habitat loss/degradation 
D Invasive species 
□ Exploitation 



lj IHnlilll 



lo ID CD co 



CO CD 



Year 



Figure X. Causes of avian extinctions over time. Totals include Extinct («=T31), Extinct in the Wild 
(n=4), and Critically Endangered (Possibly Extinct) species («=15). 



S. H. M. Butchart et al. 19 Bull. B.O.C. 2006 126A 

Plotting the pattern of the number of extinctions over time caused by the three most 
important factors (habitat loss/degradation, invasive species and exploitation) shows 
that the importance of exploitation in driving extinctions has decreased through the 
20th century whilst the importance of invasive species and habitat loss and 
degradation has increased (Fig. 8). 

Conclusions 

We developed and used the framework presented here to identify 15 Critically 
Endangered bird species as Possibly Extinct. Combining data on these species with 
data for 135 Extinct and Extinct in the Wild species shows that over the last century 
bird species have become extinct at a rate of one every 1.8 years. Habitat loss and 
degradation, invasive species and exploitation have been the main causes of 
extinction. Although the vast majority of documented extinctions thus far have been 
on islands, if we continue to degrade and destroy vast areas of natural habitats then 
it will be difficult to prevent even more extinctions from occurring imminently on 
continents. 

Acknowledgements 

For helpful discussions on interpreting the likelihood of extinction for various species we thank Nigel 
Collar, Mike Crosby, Guy Dutson and David Wege, and for comments on methodology we thank Simon 
Stuart, Janice Chanson, Georgina Mace, Craig Hilton-Taylor and Mike Hoffmann. Ana Rodrigues, 
Martin Sneary and Mike Evans kindly assisted in data extraction and analysis. For helping create Fig. 3 
we thank Mike Hoffmann and Mark Balman. We acknowledge the invaluable contribution of the 
hundreds of contributors who have provided input to the species accounts for all species maintained by 
BirdLife International in its World Bird Database, upon which these analyses are based. Simon Stuart and 
Jonathan Baillie provided helpful comments on the submitted draft. 

References: 

Atkinson, C. T., Woods, K. L., Dusek, R. J., Sileo, L. S. & Iko, W. M. 1995. Wildlife disease and conser- 
vation in Hawaii: pathogenicity of avian malaria {Plasmodium relictum) in experimentally infected 

Iiwi (Vestiaria coccinea). Parasitology 111: S59-S69. 
Balmford, A. 1996. Extinction filters and current resilience: the significance of past selection pressures 

for conservation biology. Trends Ecol. Evol. 11: 193-196. 
Bhushan, B. 1986. Rediscovery of the Jerdon's or Double-banded Courser Cursorius bitorquatus (Blyth). 

J. Bombay Nat. Hist. Soc. 83: 1-14. 
BirdLife International. 2001. Threatened birds of Asia: the BirdLife International Red Data Book. 

BirdLife International, Cambridge, UK. 
BirdLife International. 2004. Threatened birds of the world 2004. CD-ROM. BirdLife International, 

Cambridge, UK. 
Blackburn, T. M., Cassey, R, Duncan, R. R, Evans, K. L. & Gaston, K. J. 2004. Avian extinctions and 

mammalian introductions on oceanic islands. Science 305: 1955-1958. 
Brooks, T. M. 2000. Extinct species. Pp. 701-708 in BirdLife International Threatened birds of the world. 

BirdLife International, Cambridge, UK & Lynx Edicions, Barcelona. 
Clavero, M. & Garcia-Berthou, E. 2005. Invasives are a leading cause of animal extinctions. Trends Ecol. 

Evol. 20:110. 
Collar, N. J. 1998. Extinction by assumption; or, the Romeo Error on Cebu. Oryx 32: 239-244. 
Diamond, J. M. 1987. Extant unless proven extinct? Or, extinct unless proven extant? Conserv. Biol. 1: 

77-79. 



\ //. \/. Butchari et al. 20 Bull. B.O.C. 2006 126A 

Dutson, G. C. L., Magsalay, P. M. & Timmins, R. J. 1993. The rediscovery of the Cebu Flowerpecker 
Dicaeum quadricolor, with notes on other forest birds on Cebu, Philippines. Bird Conserv. Intern. 3: 
235 -243. 

Galbreath, R. & Brown, D. 2004. The tale of the lighthouse-keeper's cat: discovery and extinction of the 
Stephen's Island Wren {Traversia lyalli). Notornis 51: 193-200. 

Gurev itch, J. & Padilla, D. K. 2004. Are invasive species a major cause of extinctions? Trends Ecol. Evol. 
19:470-474. 

Hambler. C. 2004. Conservation. Cambridge Univ. Press. 

Hawkins, F., Andriamasimanana, R., Sam The Seing & Rabeony, Z. 2000. The sad story of Alaotra Grebe 
Tachybaptus rufolavatus. Bull. Afr. Bird CI. 7: 115-117. 

IUCN. 200\. IUCN Red List categories and criteria: version 3.1. IUCN SSC, Gland, & Cambridge, UK. 

IUCN. 2004. The 2004 IUCN Red List of threatened species. Available at: www.redlist.org. 

Johnson. T.H.& Stattersfield, A. J. 1990. A global review of island endemic birds. Ibis 132: 167-180. 

Juniper, T 2003. Spix's Macaw: the race to save the worlds rarest bird. London: Fourth Estate. 

Lamoreux, J., Akcakaya, H. R., Bennun, L., Collar, N. J., Boitani, L., Brackett, D., Brautigam, A., 
Brooks, T M., da Fonseca, G. A. B., Mittermeier, R. A., Rylands, A. B., Gardenfors, U., Hilton- 
Taylor, C, Mace, G, Stein, B. A. & Stuart, S. 2003. Value of the IUCN Red List. Trends Ecol. Evol. 
18:214-215. 

Magsalay, P., Brooks, T, Dutson, G & Timmins, R. 1995. Extinction and conservation on Cebu. Nature 
373: 294. 

Manne, L. L., Brooks, T. M. & Pimm, S. L. 1999. Relative risk of extinction of passerine birds on con- 
tinents and islands. Nature 399: 258-261. 

Pimm, S. L. & Brooks, T M. 1999. The sixth extinction: how large, how soon, and where? Pp. 46-62 in 
Raven, P. H. (ed.) Nature and human society: the quest for a sustainable world. National Academy 
Press, Washington DC. 

Pimm, S. L., Russell, G J., Gittleman, J. L. & Brooks, T. M. 1995. The future of biodiversity. Science 
269: 347-350. 

Pratt, T. K., Fancy, S. G & Ralph, C. J. 2001. 'Akiapola'au (Hemignathus munroi) and Nukupu'u 
(Hemignathus lucidus). In Poole, A. & Gill, F. (eds.) The birds of North America, no. 600. The Birds 
of North America, Philadelphia, PA. 

Reynolds, M. H. & Snetsinger, T J. 2001. The Hawaii rare bird search 1994-1996. Stud. Avian Biol. 22: 
133-143. 

Riley, J. & Wardill, J. C. 2001. The rediscovery of Caerulean Paradise-flycatcher Eutrichomyias rowleyi 
on Sangihe, Indonesia. Forktail 17: 45-55. 

Rodrigues, A. S. L., Pilgrim, J. D., Lamoreux, J. F., Hoffmann, M. & Brooks, T. M. (2006) The value of 
the IUCN Red List for conservation. Trends. Ecol Evol. 21: 71-76. 

Scott, J. M., Mountainspring, S., Ramsey, F L. & Kepler, C. B. 1986. Forest bird communities of the 
Hawaiian Islands: their dynamics, ecology, and conservation. Stud. Avian Biol. 9: 1-431. 

Solow, A. R. 1993. Inferring extinction from sighting data. Ecology 74: 962-964. 

Steadman, D. W. 1995. Prehistoric extinctions of Pacific island birds: biodiversity meets zooarchaeolo- 
gy. Science 267: 1123-1131. 

Tyrberg, T & Milberg, P. 1991. Xenicus lyalli: utrotad av Polynesiernas rata och fyrvaktarens katt. Vdr 
Fdgelv. 50: 15-18. 

van Riper, C, van Riper, S. G, Goff, M. L. & Laird, M. 1986. The epizootiology and ecological signif- 
icance of malaria in Hawaiian land birds. Ecol. Monogr. 56: 327-344. 

Whitten, A. J., Bishop, K. D., Nash, S. V. & Clayton, L. 1987. One or more extinctions from Sulawesi, 
Indonesia? Conserv. Biol. 1 : 42^48. 

Addresses: S. H. M. Butchart & A. J. Stattersfield, BirdLife International, Wellbrook Court, Girton Road, 
Cambridge CB3 0NA, UK. T M. Brooks, Center for Applied Biodiversity Science, Conservation 
International, 1919 M Street, NW Suite 600, Washington DC 20036, USA. 



S. H. M. Butchart et al. 



21 



Bull. B.O.C. 2006 126 A 



^ •£>£ 



tS ©.a 



S 5 a> :a » 



s3;om £ g 



p"o£ CO c2 





cu 




^g 


very 
thougl 
cates 
s may 




.. DO lO 




> 

CO 

"5 
£3 




-a 




rently reliable 
Canada in 19: 
firmed report 
nain 1990. 


CD 




o 
o 
g 

CD 
-O 


ent disco 
habitat ( 
iled) indi 
such area 








t3 


O W ,c3 ._ 




c3 _, C • H 


3 


CD 

Oh 




CJ 


23^ o3 

en co _c 


d 

s 

CD 


dn .„ y dj 


-a 


o 

GO 


-2 

CO 

c 
o 


Some t 
1990s; 
of suit; 
survey 
that ot 


Four a 
reports 
andur 
inArg 



sureuisj nniqepf 

spodaj 
psuuiiuooufi 

Ajijiqepaiap Avo^ 



sXaAjns 
ajBnbap^ui 



sXsAins 
IU909.I p00{) 

^uoipuirxg 
oi pasodsipWj 



v sasssoojd 

g §UIU9JB9JU1 9J9A9C 

S 9UI{99p JU909J 

W p9JU9Uin90p-||9y\\. 

pjoaaa 
paui-iijuoo jsbi 



Ph ■■» 



-a £ 



1 § I 



to J3 




<D o 


— «j 


T3 _0 




"S a 




Preda 
nestir 
by go 


o 




o 


CM 



.2 ■" 

'5 o 



5a" & 

J2 4^ 



OS — 



-I 

1J 

-a K 

(L> K 

If 
jl 



3 

s 

u -^ 

2 B 

to -a 

e ^ 

11 



1 s 



a 


IK 

3 


13 js 

H "3 




to 
t3 ■« 


» g 


2 


3 s 




B 

1 


8 i 

8 b 
o £ 


2 a 

W3 B 

CD p 

§"^ 


s 


-a a 

c3 cx> 




=3 8 



aj Si, 
toB 



3 cS 



u 



CO 

s 

2 s 

^ I 

CO ,3 



Oft, 



S. H. M. liuh-lhin el al. 



22 



Bull. B.O.C. 2006 126 A 



111!! 

III 5 ! 
1831.1 

c . hr rx> -9 



SJD 



B 3 



C£'is. -j: IS} C_ 



I 

> g 

"S-'E 
Si 






C3 c/3 

S 5 



eg e o — P 

ifO.O« if o. 

g-g €.3 §- 

— K en _c — 



o 
u 

«5 



■2 -2 8 - 



■O P JD j-f -P P J3 



S to <£> 

•is a 

2 -a 2 

c <u £ 

*2 -a a. 

IS g c 



c 



B.SP 



2 «S -Si jo 5 






oo; -<k o 



SI | g& 

5 2 ~~ y 2 

e3 -S '-2 ^ g 

%' 9 &.§) „«a !" 



n 


U-l 


z 


>p 


UJ 


X) 


Cu 


CO 


O- 


CO 


< 


c_ 



S3}0\[ 



SUIBIU3J JBJiqBH 



pauiiijuoouQ 



Alijiqepapp A\oq 



a, S3SS330jd 

C SUIU3JB3JU} 3J3A3C 

•"2 3inj03p JU303J 

W p3JU3Uin30p-[I3y\\ 

PJ033J 

pauuijuoa jseq 



sXsAjns ^ 



sX3Ains 

JU303J pOOQ 
(,UOipUl}X3 

oj pssodsipsjj 



00 

O <A 



— -o 



.~ & 



3 ofi -a 
p ~ -s 



o 



II 

1* E 



I § s <y §■ 



c =o 



o -5 
£ 



RJ 



«S 2 £ ~S -S "15, 



S. H. M. Butchart et al. 



23 



Bull. B.O.C. 2006 126A 



>§ 
fc.a 

C Oh 

O) co 

§1 

o.S 



^8 
« . ~ 

<U C 

S-H (L) 

Oh -a 



I* 



w o 



S3 



S-S -9 e 



■^ P "3 



o ^ 
cdfflS Q* 



t s.s 



CS O 

<u o 

00 CN 



•sl 



.£P£v 

CO ^ O 

U UN 



O IE " O 

o o > >a 

C s o « 



9 „ 



00 



S '5 .£P E 

" a co ih 

S Si-^ o 

O — r9 2 

— «? S3 E 



3 g 



6 £ 

00 m 



O rt ,,_, 



C r m 

o§.s 



si 

o Si 






CN — 



£ « 

£ £ 



§1 

co £3 






1*8 



>s 



C«.tJ « 

C o o 

■IS 
if 1 

^ P ^ 



Kj £> 



>1 

o 




fin 


c 


aj 


■5 




o 


C3 


is 



CO O — ' 



_o 



1.1 



"es -O <U 



- -a 1 
1 ? 1 
111 

£ h SC 



2 

Oh 

-a 



o ^H 



m —i 






3? ^ 3^ 



0< ^3 



o S 



2 

I m 

O co 

E ^ 



2 -5 






1 £ 



a o a 

,-s e a 



-fr§ .9 

V. to H 



ZQ, mO ZO t^O pqq: 



& E 

> J 



cd r> 

SO 



e 

oo p 

.9 b 

^ I - 

as T3 






^^ •!§ 



a 6 

3?1 



00^= 
CD - i3 



S3 gn 

Oh 2 

9 G 



^ be 



co >s O0 ^ 

8-2 Si K 

3 ^ to O 



<i: 2^ oo^ O^ US 



5. H. M. Buichart et al. 



24 



Bull. B.O.C. 2006 126A 



BB 



o 

o 
c 

—. -d 

>- o 
—. o 



u or 

8 o .2P-S 



o -d 

sl.i 

3 C S 

y * 3 

or 1 

c 

>■> <L> 
<L> X) 

3 c8 






aj o J= t2 O 



>-SF 



■a °„o 

am *; 



53 > i- 
82 "^ 
£ £-£ 

oo £ ^ 



*o -§ 



CQ ^ 

OJ O 

£ ° 

53 ™ 



00 ^ 

+-T d 



O^ 



w 5 * 3 



o £ 



s.-^ 



« (2 **, n 



d x> 

11 

00 J2 

d d 



d d 

00 13 



T3 

d 

w 
J? 
o 

'C 

o 



- d r^ 



CO 



O cd 
d'| 

•££ 

KS -q 

O <U 

CX-d 



uo D 

S§ 



| 3 1 g | 3 8 3 

filial sill 

*"" -I" .1 I X -« °« "1 ffl g 

an 5 ja, ■£ Ctq jz r O as -j 



1.1 



t5 CG 

X> -a 
DC a. 



'53 .a 



E> bo 
£r3 



,- Co £ d 
43 o -ts P 

*- ~s d u 

"§ R -° 
Xj •- ^ o 

"^ oo vo </5 

d°">§^ 



« 0J^;~ 
o «n .S 

I CM g 

O ~ 
O .2 U 

uj a s 

Q o 
< Q fe 
w «J — 
H x.' j 'n 
O d g 
ZSCQ 



tSD 



XI vo 

u o 

i- o 

s-s 

'•5 

00 OJ 

& ^ 

"^ oJ 
53 «y 

1 & 

d x 

PJ CO 



H. Glyn Young & Janet Kear 



25 



Bull. B.O.C. 2006 126 A 



The rise and fall of wildfowl of the western Indian 
Ocean and Australasia 

by H. Glyn Young & Janet Kear+ 

Wildfowl (Anseriformes) are amongst the most widespread of vertebrates. Highly 
volant, dispersive and often long-distance migrants, wildfowl have been recorded on 
almost every landmass and have colonised many of the world's islands, often 
radically changing their morphology in the process (Lack 1970, Weller 1980). 

The islands of the western Indian Ocean (Madagascar and associated islands, 
Amsterdam, and the subantarctic islands of Crozet, Kerguelen and St Paul), the 
Andamans, the Greater and Lesser Sundas, Moluccas, Philippines, New Guinea, 
Australia and New Zealand (Map 1) lie in three zoogeographical regions (the 
Malagasy, Oriental and Australasian Regions). However, whilst those wildfowl 




Figure 1 . Map of islands included in this review. 



//. Glyn Young & Janet Kear 26 Bull. B.O.C. 2006 126A 

species found there may have differing evolutionary origins, there is also evidence of 
a dispersal by colonising ancestral species (notably grey teal and white-eyed 
pochards) around the region. It is, therefore, not unreasonable to consider these 
seemingly unrelated islands together for this review of regional colonisation and 
extinction in wildfowl. 

Excluding a small number of migrant wildfowl from Eurasia, and introduced 
species, 65 taxa are known to have been resident in the region during the Holocene 
(Table 1). This list includes several taxa known only from subfossil remains 
(taxonomy and nomenclature of extant taxa is from Kear (2005) and extinct taxa 
from Livezey (1997) and Holdaway et al. (2001): see these publications and Young 
et al. (1996) for extensive reference lists). Of resident taxa, 57 (87%) are endemic to 
the region but 19 endemic wildfowl taxa (33% of regional endemics) have become 
extinct during the Holocene. All extinctions are from the Malagasy faunal region, the 
subantarctic islands and New Zealand; the precise number of extinct taxa from this 
region is still unknown (e.g. Young et al. 1996, Holdaway et al. 2001). 

Our aim here is to provide a brief overview of the diversity of wildfowl that have 
colonised the islands of the western Indian Ocean and Australasia, and their decline 
in recent times. Two taxa considered to have become extinct most recently 
(Madagascar Pochard Aythya innotata and Auckland Islands Merganser Mergus 
australis) are detailed separately. 

The rise of wildfowl in the region 

Wildfowl have dispersed freely into the region from elsewhere in the world and 
representatives of most of the popularly recognised groups have colonised the 
islands. Johnsgard's (1978) proposed system of tribes, whilst now considered overly 
simplified (see Livezey 1997, Callaghan & Harshman 2005), is still widely used and, 
of the 13 tribes listed by Johnsgard, members of 12 have been resident in the region 
in recent times (Table 2). 

The exact geographic origins of many of the resident wildfowl are often unclear. 
Many taxa, especially those in Australasia, have no close living relatives outside the 
region. Wildfowl are probably southern and tropical in origin (Kear 1970, Callaghan 
& Harshman 2005) and it is perhaps unsurprising that so many ancient and 
distinctive taxa are known from the region. For example, the Australian Magpie 
Goose Anseranas semipalmatus, sole member of the Anseranatidae, and New 
Zealand's river specialist, the Blue Duck Hymenolaimus malacorhynchos, which is 
typically placed in a grouping with the South American Torrent Duck Merganetta 
armata (Callaghan & Harshman 2005), have no obvious relatives anywhere (see 
Callaghan 2005a). Two further, aberrant, sister genera, Cnemiornis (two species) in 
New Zealand and Cereopsis in Australia are related most closely to another curious 
species, the South American Coscoroba Swan Coscoroba coscoroba (St John et al. 
2005). 

New Guinea's Salvadori's Duck Salvadorina waigiuensis and the Pink-eared 
Ducks (Malacorhynchus membranaceus in Australia and M. scarletti in New 



H. Glyn Young & Janet Kear 



27 



Bull. B.O.C. 2006 126A 



TABLE 1 
Wildfowl species resident in western Indian Ocean and Australasia. Taxa endemic to region are in bold, 

extinct taxa are underlined; ▲ taxon resident; G taxon extinct. NZ New Zealand, AUS Australia, 

NG New Guinea, PH Philippines, MOL Moluccas, SUND Sundas, AND Andaman and Nicobar Islands, 

MAD Malagasy Faunal Region (Madagascar, Comores, Seychelles & Mascarene Islands), 

SUB Subantarctic islands (Amsterdam, Crozet, Kerguelen and St Paul). 



Species 

Magpie Goose Anseranas semipalmatus A A 

White-faced Whistling-duck Dendrocygna viduata 
Spotted Whistling-duck Dendrocygna guttata A 

Fulvous Whistling-duck Dendrocygna bicolor 
Plumed Whistling-duck Dendrocygna eytoni A 

Wandering Whistling-duck Dendrocygna arcuata australis A 

Dendrocygna a. arcuata A 

Dendrocygna a. pygmaea A 

Lesser Whistling-duck Dendrocygna javanica 
White-backed Duck Thalassornis leuconotus insularis 
Musk Duck Biziura lobata A 

New Zealand Musk Duck Biziura delatouri D 

Black Swan Cygnus atratus D A 

Cape Barren Goose Cereopsis n. novaehollandiae A 

Cereopsis novaehollandiae grisea A 

South Island Goose Cnemiornis calcitrans □ 

North Island Goose Cnemiornis gracilis D 

Freckled Duck Stictonetta naevosa A 

Blue-billed Duck Oxyura australis A 

Blue Duck Hymenolaimus malacorhynchos A 

African Comb Duck Sarkidiornis melanotos 
Greater Madagascan Sheldgoose Centrornis majori 
Lsr. Madagascan Sheldgoose Alopochen sirabensis 
Mauritius Sheldgoose Alopochen mauritianus 
Reunion Sheldgoose Alopochen kervazoi 
Chatham Island Duck Pachyanas chathamica D 

Radjah Shelduck Tadorna radjah rufitergum A 

Tadorna radjah radjah A 

Australian Shelduck Tadorna tadornoides A 

Paradise Shelduck Tadorna variegata A 

Chatham Shelduck Tadorna sp. G 

Pink-eared Duck Malacorhynchus membranaceaus A 

Scarlett's Duck Malacorhynchus scarletti 3 

Salvadori's Duck Salvadorina waigiuensis A 

White-winged Duck Asarcornis scutulata leucoptera 
Australian Wood Duck Chenonetta jubata A 

Finsch's Duck Chenonetta finschi D 

African Pygmy-goose Nettapus auritus 
Cotton Teal Nettapus c. coromandelianus 

Nettapus c. albipennis A 

Green Pygmy-goose Nettapus pulchelus A A 



NZ AUS NG PH SUND MOL AND MAD SUB 



//. Glyn Young & Janet Kear 28 Bull. B.O.C. 2006 126A 

Amsterdam Island Duck Anas marecula D 

St Paul Island Duck Anas sp. □ 

Philippine Duck Anas luzonica A 

Meller's Duck Anas melleri A 

Australasian Shoveler Anas rhynchotis A A 

Madagascar Teal Anas bernieri A 

Sauzier's Teal Anas theodoh 3 

Indonesian Teal Anas gibberifrons A A 

Grey Teal Anas gracilis ▲ ▲ ▲ 

Andaman Teal Anas albogulahs A 

Chestnut Teal Anas castanea A 

Brown Teal Anas chlorotis A 

Auckland Island Teal Anas aueklandica A 

Campbell Island Teal Anas nesiotis D 

Macquarie Island Teal Anas sp. ▲ 

Red-billed Pintail Anas eiythrorhyncha A 

Kerguelen Pintail Anas eatoni eatoni A 

Crozet Pintail Anas eatoni drygalskii A 

Hottentot Teal Anas hottentota A 

Hardhead Aythya australis A 

Madagascar Pochard Aythya innotata D 

'Reunion Pochard' Aythya sp. G 

New Zealand Scaup Aythya novaeseelandiae A 

Auckland Islands Merganser Mergus australis O 

TABLE 2 

Wildfowl tribes represented in western Indian Ocean, Australasia, South America and Africa. 

* Johnsgard (1978) placed Blue Duck Hymenolaimus malacorhynchos in Anatini. 



Tribe (from Johnsgard 1978) 


Tribe in 


Tribe in 


Tribe in 


Tribe in 


Tribe in 




west 


Africa 


Andamans, 


Australasia 


South 




Indian 


(Brown 


Sundas, 


(this study) 


America 




Ocean 


etal. 


Moluccas and 




(Blake 




(this study) 


1982) 


Philippines 
(this study) 




1977) 


Anseranatini (Magpie Goose) 








/ 




Dendrocygnini (Whistling-ducks) 


/ 


/ 


/ 


/ 


/ 


Anserini (Swans and True Geese) 








/ 


/ 


Cereopsini (Cape Barren Goose) 








/ 




Stictonettini (Freckled Duck) 








/ 




Tadornini (Shelducks and Sheldgeese) 


/ 


/ 


/ 


/ 


/ 


Tachyerini (Steamer Ducks) 










/ 


Cairinini (Perching Ducks) 


/ 


/ 


/ 


/ 


/ 


Merganettini (Blue and Torrent Duck)* 








/ 


/ 


Anatini (River Ducks) 


/ 


/ 


/ 


/ 


/ 


Aythyini (Pochards) 


/ 


/ 




/ 


/ 


Mergini (Sea Ducks) 








/ 


/ 


Oxyurini (Stiff-tailed Ducks) 


/ 


/ 




/ 


/ 



H. Glyn Young & Janet Kear 29 Bull. B.O.C. 2006 126A 

Zealand) have also proven difficult to place (see Callaghan 2005b). These very 
distinctive ducks generally now comprise their own tribe (Malacorhynchini), an early 
branch of the true ducks (subfamily Anatinae); however, this placement is very 
tentative, as, possibly, is the inferred relationship between the two genera. 

There are also representatives in the region of larger 'modern' wildfowl tribes that 
have no extant relatives in adjacent continental areas (e.g. Cygnus, Mergus and 
Oxyurd). The absence of these genera from the nearby mainland further suggests 
early radiations in the region during previous geological periods, when wildfowl 
habitats may have been different. As a consequence of sea level rises following the 
Pleistocene, a diverse range of taxa from both ancient and modern wildfowl groups 
has evolved on islands. 

Taxonomic revisions have frequently changed the status of island forms. The 
Madagascan Meller's Duck Anas melleri, once considered a poorly differentiated 
island isolate of the Holarctic Northern Mallards, platyrhynchos is currently deemed 
a much older species with close affinities to African ducks (Young & Rhymer 1998). 
The extinct Amsterdam Island Duck A. marecula (see below) was initially thought 
derived from the highly dispersive Palearctic Garganey A. querquedula (Bourne et al. 
1983, Martinez 1987), until closer examination suggested that the species was a form 
of wigeon (Olson & Jouventin 1996). 

Resident populations of the widespread whistling-ducks Dendrocygna viduata 
and D. bicolor have become established in Madagascar and are undifferentiated from 
those elsewhere in the world. The presence of so many whistling-duck taxa in the 
region (six of eight species of Dendrocygna and the single species of Thalassornis; 
three of them endemic species) is interesting, although its significance in the 
evolution of this subfamily is unclear (Livezey 1995). There is also a high number of 
species of Tadornini (shelducks and sheldgeese) in the region. The limits of this 
group are contentious, however, as the five genera recognised by Callaghan & 
Harshman (2005; including three recently extinct Alopochen species from the 
Malagasy Region) and the poorly known subfossil genera Centrornis (Madagascar: 
Livezey 1997, Goodman & Hawkins 2003) and Pachyanas (New Zealand: Worthy 
& Holdaway 2002) include 20 species, of which nine occur in the region compared 
with six in South America and four in Africa. 

The African origin of many wildfowl in Madagascar is unsurprising and three 
Afrotropical species (Anas erythrorhyncha, A. hottentota and Nettapus auritus) and 
Sarkidiornis melanotos (found in the Afrotropical and Oriental Regions, with a sister 
species, S. sylvicola, in South America) are widespread and resident throughout 
Madagascar. Migration between Madagascar and mainland Africa has never been 
observed, but has been proposed for at least one species (A. erythrorhyncha: 
Langrand 1990). The endemic Madagascar White-backed Duck Thalassornis 
leuconotus insularis is a small, dark, form of this Afrotropical species (Young 2005) 
and the endemic Meller's Duck, undoubtedly a long-established species in 
Madagascar, appears to have shared-ancestry with two African species, Black Duck 
A. sparsa and Yellow-billed Duck A undulata (Young & Rhymer 1998). Four extinct 



//. Gfyn Young & Janet Kear 30 Bull. B.O.C. 2006 126A 

shcldgeese are known from the Malagasy Region (Centrornis majori and Alopochen 
sirabensis in Madagascar; A. mauritianus in Mauritius; and A kervazoi in Reunion), 
and those placed to date in Alopochen are presumably closely related to African A. 
aegyptiacus (Livezey 1997). 

Surprisingly, with many islands in the region lying close to the Asian mainland, 
only a small number of continental species (Radjah Shelduck Tadorna radjah, 
White-winged Duck Asacornis scutulata and Cotton Teal Nettapus 
coromandelianus) have forms in the region, and there is also only one, 
undifferentiated, wildfowl species (Lesser Whistling-duck D. javanica) resident. 
Mallards are represented in the Philippines by Philippine Duck Anas luzonica and in 
Australasia by Pacific Black (Grey) Duck A. superciliosa which are, with the 
mainland's spot-billed ducks A. poecilorhyncha and A zonorhyncha, members of an 
Asian subgroup within the mallard clade (Livezey 1997, Johnson & Sorenson 1999). 
The Australian pochard, Hardhead Aythya australis, is a member of an Aythya clade 
(the white-eyed pochards) with the Ferruginous Duck ,4. nyroca and Baer's Pochard 
A. baeri representing Asian populations, whilst New Zealand's scaup A. 
novaeseelandiae is a member of an entirely different pochard clade (the scaup), 
which includes Greater Scaup A. marila, Lesser Scaup A. americana and Tufted 
Duck A. fuligula. The two Australasian pochards, Hardhead and New Zealand Scaup, 
have quite distinct origins (Livezey 1997). Madagascar Pochard A. innotata (see 
below) is another white-eyed pochard that may be more closely related to the 
Australian species. Subfossil remains of an Aythya in Reunion have been tentatively 
linked to A. innotata but may be a distinct taxon (Mourer-Chauvire et al. 1999). 

Blue-billed Duck Oxyura australis is part of a recent clade within the stifftail 
tribe (Oxyurini), which includes Afrotropical O. maccoa and Palearctic O. 
leucocephala (McCracken & Sorenson 2004) (note, the extinct New Zealand 
population of O. australis (see Horn 1983) is now considered a misidentification: 
Holdaway e/ a/. 2001). 

Palearctic affinities in the region's wildfowl are most obvious on the subantarctic 
islands but representatives of several genera such as swans (Cygnus), pochards 
(Aythya), stifftails (Oxyura) and mergansers (Mergus) today have closest relatives in 
the Palearctic. The southern taxa, at the same time as their current northern 
counterparts, may have evolved following isolation in the Southern Hemisphere. The 
number of swan species in the region has been debated. Black Swan C. atratus is 
common in Australia and New Zealand, and migrants have repopulated areas where 
they were exterminated by Polynesian settlers (Holdaway et al. 2001, Williams 
2003). Further introductions have occurred following the arrival of Europeans. New 
Zealand's extinct swan population was originally described, from subfossil remains, 
as C. sumnerensis (Livezey 1989, Turbott 1990), but separation from C. atratus is no 
longer accepted (Worthy & Holdaway 2002). Another extinct swan, which once 
occurred on the Chatham Islands, is awaiting analysis and may be distinct from the 
Black Swan (Holdaway et al. 2001). 



H. Glyn Young & Janet Kear 3 1 Bull. B. O. C. 2006 1 26A 

The ducks of the remote subantarctic islands owe their origins to a variety of 
source species. The pintail taxa, Eaton's Pintail Anas eatoni, resident on Kerguelen 
(A. eatoni eatoni) and Crozet (A. eatoni drygalskii) are considered to be forms of the 
Holarctic Northern Pintail A. acuta (Marchant & Higgins 1990). The extinct 
flightless ducks of Amsterdam and St Paul {A. marecula and an undescribed taxon, 
respectively) are probably forms of wigeon (three species of wigeon occur outside 
the region). Interestingly, whilst the undescribed (and potentially flightless) duck 
from St Paul may be related to and share a common ancestor with A. marecula, it is 
also possible that it represents a different taxon (Olson & Jouventin 1996). 

Brown teals are known from New Zealand (A. chlorotis) and three isolated 
subantarctic island groups: the Aucklands {A. aucklandica), Campbells (A. nesiotis) 
and Macquarie (undescribed taxon, see Holdaway et al. 2001), all of which islands 
are included within New Zealand for the purposes of this publication. A. aucklandica 
and A. nesiotis are flightless (flightlessness is indeterminate in the Macquarie species, 
but would be expected). This radiation is separate from the similar diversification 
within the related grey teal (see below), and the ancestral brown teal most likely 
reached New Zealand well before the modern radiation of grey teal (Daugherty et al. 
1999, Kennedy & Spencer 2000). 

Dispersal of taxa within the region 

There is significant evidence to show that some wildfowl have dispersed within the 
region, without obvious additional contact with continental areas. Grey teal taxa are 
resident in Australia (Chestnut Teal A. castanea), Australia, New Zealand, New 
Guinea and New Caledonia (Grey Teal A. gracilis), the Sundas and Moluccas 
(Indonesian Teal A. gibberifrons), Andamans (Andaman Teal A. albogularis) and 
Madagascar (Madagascar Teal. A. bernieri). Populations of grey teal are still often 
highly dispersive. As these species are adapted to saline waters and breed in 
mangrove habitat, a possible pattern of dispersal by a common ancestor during 
Pleistocene sea level changes is postulated (Young 2002), particularly as lower sea 
levels would have permitted the development of extensive mangroves. The extinct 
duck from the Mascarenes, Sauzier's Teal A. theodori was probably also a grey teal 
(Mourer-Chauvire et al. 1999) and other taxa in this clade may have been resident in 
areas now submerged. 

The wildfowl fauna of New Zealand provides further evidence of movement 
within the region, revealing an obvious connection with Australia; many forms have 
presumably colonised directly from the larger island. Of 18 recent species (Table 1), 
four (Cygnus atratus, Anas superciliosa, A. gracilis and A. rhynchotis) have 
populations in both islands and a further seven (Cnemiornis (two species, with 
Cereopsis), Tadorna (two species including the undescribed Chatham Island 
Shelduck), Chenonetta (Euryanas), Malacorhynchus and Biziura have counterparts 
in Australia. The four brown teals probably also evolved from a common ancestor of 
this group and the grey teal that reached New Zealand from Australia. 



//. Gfyn Young & Janet Kear 32 Bull. B.O.C. 2006 126A 

The fall 

The decline of the region's wildfowl during recent times follows an ail-too 
predictable course and one that has been thoroughly documented elsewhere (see e.g. 
Olson & James 1984, Milberg & Tyrberg 1993, Steadman & Martin 2003). Typically, 
the discovery and subsequent colonisation of the more remote islands by humans 
heralded the decline of their wildfowl faunas. The exact agents of extinction on 
specific islands are sometimes debatable but habitat modification, predation from and 
competition with exotic animals and direct persecution, are all generally 
anthropogenic in source. 

In New Zealand, 18 species of wildfowl were resident at first human contact 
(Holdaway et al. 2001). Not included in this number is Anas rhynchotis, as this 
species is considered to have immigrated naturally since human colonisation 
c. 1,000-1,200 years ago (Cassels 1984). The first Polynesian settlers colonised New 
Zealand around 800 years ago and they and their commensal mammals initiated a 
cycle of faunal extinctions throughout the archipelago that has resulted in nine 
wildfowl species (50%) becoming extinct (see Worthy & Holdaway 2002 for details 
of the decline of New Zealand's fauna and list of references). 

The Chatham Islands (New Zealand) were first colonised by Polynesians from 
New Zealand c.400^450 years ago (Tennyson & Millener 1994) and extinctions of 
wildfowl populations may have occurred later than on the main New Zealand islands. 
Two species, however, disappeared following the arrival of Europeans, the merganser 
(see below) and, presumably, the endemic teal of Macquarie. Macquarie was not 
discovered until 1810, after which it became a base for visiting sealers (Taylor 1979) 
who presumably exterminated this small insular duck. It was probably also the 
activities of sealers and commensal mammals that exterminated the Amsterdam and 
St Paul ducks (possibly as late as 1877), following their discovery and occupation by 
sealers from the end of the 17th century (Bourne et al. 1983, Jouventin 1994). 

Similarly, in the western Indian Ocean, extirpation of all native wildfowl in the 
Mascarenes followed human colonisation early in the 17th century, and all species 
were extinct by the end of the same century (Cheke 1987). In Madagascar, however, 
the extinction date of the sheldgeese is unclear. Madagascar has recently undergone 
a reduction in annual rainfall, being distinctly wetter even just 1,000 years ago 
(Hawkins & Goodman 2003). Both Centrornis majori and Alopochen sirabensis 
were once common in the south-west and the central highlands (Young et al. 2003), 
thus aridification, coupled with human agencies following colonisation, may have 
resulted in the extinction of the sheldgeese. 

Two recently extinct species are detailed below. 

Madagascar Pochard Aythya innotata 

The Madagascar Pochard (or Madagascar White-eye) superficially resembles the 
Palearctic Ferruginous Duck (or Eurasian White-eye), a species that winters in part 
in Africa and has been recorded in Seychelles (Skerrett 1999). Analysis of 



n. (jiyn loung & Janet Kear a hull h.u.i^. zuuo izoa 

morphological (Livezey 1996) and genetic (Sorenson & Fleischer 1996) material has 
revealed that the Madagascan species, whilst nestling within a distinct clade of 
Aythya that includes A. nyroca, is possibly most closely related to the Australian 
endemic Hardhead (or Australasian White-eye). 

A. innotata was described from Lake Alaotra, an extensive shallow wetland on 
the Central Plateau (c. 1,200 m above sea level) in 1894 (Salvadori 1894), and 
appears to have been restricted to this site, at least since its discovery. Supposed 
sightings away from Lake Alaotra (e.g. at Lake Ambohibao, near Antananarivo; 
Salvan 1970) are doubtful or at least must have represented rare dispersal. Subfossil 
remains of an Aythya are available from Reunion (Mourer-Chauvire et al. 1999, 
Hawkins & Goodman 2003) and these may represent dispersal events or the presence 
on this island of another, closely related species. 

The open water of Alaotra may have always been little used by pochards, as they 
were probably inhabitants of quiet, well-vegetated pools much like A. nyroca 
(Callaghan & Green 2005). Large areas of the Alaotra basin were once occupied by 
papyrus and Phragmites marsh, most notably in the south. This marsh, though much 
reduced today, is still typified by numerous quiet pools and abundant emergent 
vegetation including water lilies Nymphea sp. The wetland also once held another 
endemic waterbird, the presumably extinct Alaotra (or Delacour's) Grebe 
Tachybaptus rufolavatus and still hosts a population of marsh-living lemurs 
(Hapalemur griseus alaotrensis) that feed exclusively on marsh vegetation. 

Examination of the lake system at Alaotra today shows the area to be much 
reduced from its former extent and, whilst this reduction has been mostly 
undocumented, it must be assumed that the reduction is largely anthropogenic in 
origin. The majority of surrounding hillsides have been entirely deforested and 
resultant siltation has considerably reduced water depth and aquatic biodiversity. 
Humans first constructed settlements in Madagascar c. 1,200 years ago (Dewar 2003) 
and habitat modification throughout the island has subsequently been extensive. The 
pochard was described as common at Alaotra in 1929 (Delacour 1954) and 1935 
(Webb 1936), and was still present in 1960 (Dee 1986), when the last known 
sightings were made at the lake. Although described as common in 1978 (Soothill & 
Whitehead 1978), the species had not in fact been found during several surveys at the 
lake since 1971 (Dee 1986, Young & Smith 1989, Wilme 1994). Salvan (1970) 
provided the last published sighting anywhere, in 1970, away from Alaotra. 

Surprisingly, following a publicity campaign amongst villages surrounding 
Alaotra in 1989, a male was captured by fishermen and taken into captivity in 1991 
(Wilme 1993). Following this, further extensive surveys of Alaotra and adjacent 
wetlands on the Central Plateau failed to locate more birds (Pidgeon 1996). The 
captured bird, which died in 1992, is the last known and only photographed 
individual, and A. innotata is now considered extinct. 

The rapid decline of A. innotata went almost unnoticed, making it now difficult 
to determine the precise causes. All wetlands in Madagascar have undergone severe 
anthropogenic modification (Young 1996). These changes, however, probably 



//. Gfyn Young & Janet Kear 34 Bull. B.O.C. 2006 126A 

commenced subsequent to a lengthy period of natural aridification in parts of the 
island such as the south-west (Goodman & Rakotozafy 1997). The primary factor 
attributed to the extinction of the pochard is the introduction into Lake Alaotra (and 
many other wetlands) of exotic fish. The release of herbivorous cichlids into the lake 
(e.g. Oreochromis mossambicus, O. niloticus, O. macrochir, Tilapia rendalli and T. 
melanopleura in 1955-60; Pigeon 1996) has had considerable affect on the aquatic 
flora and entire lake system. The subsequent introduction of alien carnivorous 
species, Black Bass Micropterus salmoides and Asian Snakehead Ophicephalus 
striates, in 1961 and around 1980, respectively (Pigeon 1996), undoubtedly proved 
detrimental to the lake's waterbirds through both competition and direct predation 
(the introduction of M. salmoides into Lake Atitlan in Guatemala in 1960 has been 
directly linked to the decline and extinction of the grebe Podilymbus gigas endemic 
to this single lake: BirdLife International 2000). Increasing use by fishermen of 
monofilament gill nets that catch all unsuspecting diving birds, and pesticide run-off 
from adjacent rice fields, have placed additional pressures on the lake's waterbirds 
and probably also caused the extinction of the Alaotra Grebe (Pigeon 1996, Young 
1996, Hawkins et al. 2000), and may prevent any recolonisation by pochards, or 
grebes, if an unknown population exists elsewhere in Madagascar. 

Auckland Islands Merganser Mergus australis 

The presence of a merganser in New Zealand and its outlying islands is something of 
a biogeographic enigma, as is the bird's phylogenetic relationships. Only two 
Mergini have been identified as having Holocene distributions in the Southern 
Hemisphere but, according to Livezey (1989b), Auckland Islands Merganser Mergus 
australis and Brazilian Merganser M. octosetaceus are not close relatives; the former 
representing an early and unique branch in the merganser clade, the latter being 
closely related to Northern Hemisphere M. serrator and M. squamatus. 

At the time of first human contact, c.800 years ago, M. australis was apparently 
distributed around the coasts of New Zealand's three main islands (North, South and 
Stewart), was a member of the extensive waterfowl fauna in the Chatham archipelago 
(650 km east of New Zealand), and occurred on the subantarctic Auckland Islands 
(400 km south of New Zealand) (Millener 1999, Worthy & Holdaway 2002). 
McCormick's (1842) claim of their presence on subantarctic Campbell (600 km 
south of New Zealand) is considered a misidentification. Mergus bones have been 
found in numerous coastal fossil deposits, particularly at the heads of large and 
sheltered bays, from the north-eastern tip of North Island to Stewart Island, and in 
middens at some river mouths and estuaries in all three main islands (Worthy & 
Holdaway 2002). On Chatham, Mergus bones are common in fossil and midden 
deposits especially fringing the extensive saline Te Whanga Lagoon (Millener 1999). 
Within the Auckland group, subfossil bones have been found at the heads of some 
sheltered eastern inlets (Kear & Scarlett 1970). 

The Chatham Islands, which prior to sea level rises at the end of the Pleistocene 
was a large and extensive archipelago, had a rich avifauna of at least 100 species 



H. Glyn Young & Janet Kear 35 Bull. B.O.C. 2006 126 A 

(Millener 1996, 1999) and many species, clearly of New Zealand origin, have 
differentiated and are recognised as unique taxa (Turbott 1990). Millener (1999) 
considered the Chatham population of Mergus to represent an undescribed species, 
being smaller than the Aucklands population, with a shorter bill and reduced wings. 
This distinction, however, was not supported by Worthy & Holdaway (2002), who 
pointed to the considerable size variation evident in Chatham and mainland New 
Zealand populations. 

The early Polynesian settlers of New Zealand significantly altered the faunal 
composition of the islands, producing what has become among the best documented 
of recent human-induced extinction events (Worthy & Holdaway 2002). Mergus was 
one of eight waterfowl species exterminated within, perhaps, 300 years of the 
settlement of New Zealand, and amongst six waterfowl species subsequently 
exterminated on the Chathams when this archipelago too was settled (Millener 1999). 

Historic accounts of Auckland Islands Merganser are restricted to its distribution 
within the Auckland group (Kear & Scarlett 1970). The first specimen to be collected 
for science was in 1840, in Laurie Harbour, a northern inlet, but all subsequent 
sightings appear to have been made within the confines of the southern Carnley 
Harbour. The assiduous collector, W.L. Buller (1891) commented Tt is very desirable 
that specimens of this interesting form in the adult state should be obtained for our 
museums before it is too late', further pointing out that 'Although ....(during) 
periodical visits to Auckland Islands. . .eager search is made, the bird is scarcely ever 
seen'. Thus, the Auckland Islands Merganser was sought-after by subsequent 
visitors, along with another endemic waterfowl, the flightless Auckland Island Teal 
Anas aucklandica. Kear & Scarlett (1970) reported the whereabouts of 26 specimens, 
20 of which were collected post-Buller's 1891 remarks and the last two were 
apparently shot on 9 January 1902. As seven specimens were reputedly taken in 1901 
alone (Kear & Scarlett 1970), the possibility exists that collecting was the ultimate 
cause of its extinction. 

Few reports were made on the ecology of this species. No nest was ever recorded 
and there is only one written observation of the young (a brood of four, which were 
collected; Chapman 1891). Confusion also surrounds its primary habitat — was it a 
sea duck or mostly an estuarine and freshwater inhabitant? Kear & Scarlett (1970) 
collated all known references and reported a rare observation of birds in a small, 
steep-flowing stream. However, all other observations in Carnley Harbour appear to 
have been at the heads of sheltered bays near the emergence of freshwater streams. 
15 N and 13 C isotope signatures of two fossil bones from separate New Zealand sites 
suggest both freshwater and marine foods (R. Holdaway, M. Williams unpubl. data). 

The other islands 

Interestingly, whilst the fall of the Anseriformes in the western Indian Ocean, New 
Zealand and subantarctic islands has been dramatic, the other island groups in the 
region have seen no recent wildfowl extinctions. In Australia, megafaunal collapse 
following the arrival of humans c.50,000 years ago is well documented (Miller et al. 



//. Gfyn Young <A; Janet Kear 



36 



Bull. B.O.C. 2006 126A 



TABLE 3 

Stains o\' resident w ildfovt I taxa in island groups of the western Indian Ocean and Australasia. 

Madagascar includes islands of the Malagasy Faunal Region (see Table 1), Subantarctic Islands 

includes Amsterdam, Crozet, Kerguelen and St Paul. 

[UCN status is from TWSG (2004) and includes Critical, Endangered and Vulnerable. 



Island group 


No. of recent taxa 


No. of extinct taxa 


No. of endangered taxa 




(see 


Table 1) 


(see Table 1) 


IUCN status 
(TWSG 2003) 


Now Zealand 


18 




9 


4 


Australia 


19 








New Guinea 


9 






1 


Philippines 


3 






1 


Sundas 


5 






1 


Moluccas 


4 








Andaman Islands 


2 






1 


Madagascar 


16 




7 


3 


Subantarctic islands 


4 




2 


2 



2005) and extinctions include a possible Anseriform (Dromornis sp.; Vickers-Rich & 
Rich 1999). This may be due to the adaptation of wildfowl to human presence on 
islands closer to mainland areas, or that the remains of fossil taxa have yet to be 
discovered. The current status of the remaining taxa in the region, however, should 
not give rise to complacency — of 38 endemic species, 13 (34%) are listed by IUCN 
as Critical, Endangered or Vulnerable (TWSG 2003; Table 3). 

Acknowledgements 

We thank Julian Hume and the BOU for inviting us to give this presentation at the Linnean Society, and 
to Steve Dudley for helping with arrangements. Michael Sorenson and Murray Williams provided 
encouragement and helped with unpublished material, Roger Safford and Trevor Worthy assisted further 
in the provision of references. Tim Flannery, Peter Schouten and Julian Hume permitted use of their 
artwork in the presentation and thanks are further due to Kevin McCracken for his support. Julian Hume, 
Anthony Cheke and Guy Kirwan made useful comments on earlier drafts of this manuscript. 

Janet Kear became ill shortly before the presentation in London and was unable to attend it, passing 
away later in the same month. Janet, however, was very excited about the event and discussed by telephone 
her contribution two days before. HGY would like to thank Tim Davies and Janet's husband, John Turner, 
for support and to recognise the attendees of the symposium who expressed their best wishes on the day: 
it would not have been possible to present this discussion on the remarkable wildfowl of this region 
without their combined help. Janet intended specifically to cover the Auckland Islands Merganser, a bird 
that had long interested her, and extreme gratitude is due to Murray Williams, who wrote this section for 
her. 



References: 

BirdLife International. 2000. Threatened birds of the world. BirdLife International, Cambridge, UK & 

Lynx Edicions, Barcelona. 
Blake, E. R. 1977. Manual of Neotropical birds, vol. 1. Univ. of Chicago Press. 
Bourne, W. R. P., David, A. C. F. & Jouanin, C. 1983. Probable Garganey on St. Paul and Amsterdam 

Islands, Indian Ocean. Wildfowl 34: 127-129. 



H. Glyn Young & Janet Kear 3 7 Bull. B. O. C. 2006 1 26 A 

Brown, L. H., Urban, E. K. & Newman, K. 1982. The birds of Africa, vol. 1. Academic Press, London. 

Buller, W. L. 1891. Notes and observations on New Zealand birds. Trans. & Proc. New Zealand Inst. 24: 
64-91. 

Callaghan, D. 2005a. Magpie Goose: taxonomy. Pp. 182-183 in Kear, J. (ed.) Ducks, geese and swans. 
Oxford Univ. Press. 

Callaghan, D. 2005b. Blue Duck, Torrent Duck and steamer-ducks: taxonomy. P. 370 in Kear, J. (ed.) 
Ducks, geese and swans. Oxford Univ. Press. 

Callaghan, D. 2005c. The true ducks, subfamily Anatinae. Pink-eared Duck and Salvadori's Duck: taxon- 
omy. P. 441 in Kear, J. (ed.) Ducks, geese and swans. Oxford Univ. Press. 

Callaghan, D. & Green, A. J. 2005. Ferruginous Duck Aythya nyroca. Pp. 659-662 in Kear, J. (ed.) Ducks, 
geese and swans. Oxford Univ. Press. 

Callaghan, D. & Harshman, J. 2005. Taxonomy and systematics. Pp. 14-26 in Kear, J. (ed.) Ducks, geese 
and swans. Oxford Univ. Press. 

Cassels, R. 1 984. The role of prehistoric man in the faunal extinctions of New Zealand and other Pacific 
islands. Pp. 741-767 in Martin, P. S. & Klein, R. G. (eds.) Quaternary extinctions: a prehistoric rev- 
olution. Univ. of Arizona Press, Tucson. 

Chapman, F. R. 1891. The outlying islands south of New Zealand. Trans. & Proc. New Zealand Inst. 23: 
491-522. 

Cheke, A. S. 1987. An ecological history of the Mascarene Islands, with particular reference to extinctions 
and introductions of land vertebrates. Pp. 5-89 in Diamond, A. W. (ed.) Studies of Mascarene Island 
birds. Cambridge Univ. Press. 

Daugherty, C. H., Williams, M. & Hay, J. M. 1999. Genetic differentiation, taxonomy and conservation of 
Australasian teals Anas spp. Bird Conserv. Intern. 9: 29-42. 

Dee, T J. 1986. The endemic birds of Madagascar. International Council of Bird Preservation, Cambridge, 
UK. 

Delacour, J. 1956. The waterfowl of the world, vol. 2. Country Life, London. 

Dewar. R. E. 2003. Relationship between human ecological pressure and the vertebrate extinctions. Pp. 
119-122 in Goodman, S. M. & Benstead, J. P. (eds.) The natural history of Madagascar. Univ. of 
Chicago Press. 

Goodman, S. M. & Rakotozafy, L. M. A. 1997. Subfossil birds from coastal sites in western and south- 
western Madagascar: a paleo-environmental reconstruction. Pp. 257-279 in Goodman, S. M. & 
Patterson, B. D. (eds.) Natural and human-induced change in Madagascar. Smithsonian Institution 
Press, Washington DC. 

Hawkins, A. F. A. & Goodman, S. M. 2003. Introduction to the birds. Pp. 1019-1044 in Goodman, S. M. 
& Benstead, J. P. (eds.) The natural history of Madagascar. Univ. of Chicago Press. 

Hawkins, F., Andriamasimanana, R., The Seing, S. & Rabeony, Z. 2000. The sad story of Alaotra Grebe 
Tachybaptus rufolavatus. Bull. Afr. Bird CI. 7: 115-117. 

Holdaway, R. N., Worthy, T H. & Tennyson, A. J. D. 2001. A working list of breeding bird species of the 
New Zealand region at first human contact. New Zealand J. Zool. 28: 119-187. 

Horn, P. L. 1983. Subfossil avian deposits from Poukawa, Hawkes Bay, and the first record of Oxyura aus- 
tralis (Blue-billed Duck) from New Zealand. J. Roy. Soc. New Zealand 13: 67-78. 

Johnsgard, P. A. 1978. Ducks, geese, and swans of the world. Univ. of Nebraska Press, Lincoln. 

Johnson, K. P. & Sorenson, M. D. 1999. Phylogeny and biogeography of dabbling ducks (genus: Anas): a 
comparison of molecular and morphological evidence. Auk 116: 792-805. 

Jouventin, P. 1994. Past, present and future of Amsterdam Island (Indian Ocean) and its avifauna. Pp. 
122-132 in Nettleship, D. N., Burger, J. & Gochfeld, M. (eds.) Seabirds on islands: threats, case stud- 
ies and action plans. BirdLife International, Cambridge, UK. 

Kear, J. 1970. The adaptive radiation of parental care in waterfowl. Pp. 357-392 in Crook, J. H. (ed.) 
Social behaviour in birds and mammals. Academic Press, London. 

Kear, J. (ed.) 2005. Ducks, geese and swans. Oxford Univ. Press. 

Kear, J. 2005. Auckland Islands Merganser Mergus australis. Pp. 746-746 in Kear, J. (ed.) Ducks, geese 
and swans. Oxford Univ. Press. 



//. Gfyn Young & Janet Kear 38 Bull. B.O.C. 2006 126A 

Kear, J. & Scarlett. R. J. 1970. The Auckland Islands Merganser. Wildfowl 21: 78-86. 

Kennedy, M. & Spencer, H. G 2000. Phylogeny, biogeography, and taxonomy of Australasian teals. Auk 
117: 154-163. 

Lack. D. 1970. The endemic ducks of remote islands. Wildfowl 19: 67-69. 

1 ivezey, B. C. 1989a. Phylogenetic relationships of several subfossil Anseriformes of New Zealand. Occ. 
Papers Mus. Nat. Hist., Univ. Kansas 128: 1-25. 

Livezey. B. C. 1989b. Phylogenetic relationships and incipient flightlessness of the extinct Auckland 
Islands Merganser. Wilson Bull. 101:410-435. 

Livezey, B. C. 1996. A phylogenetic analysis of modern pochards (Anatidae: Aythyini). Auk 113: 74-93. 

Livezey, B. C. 1997. A phylogenetic classification of waterfowl (Aves: Anseriformes), including selected 
fossil species. Ann. Carnegie Mus. 66: 457^96. 

Marchant. S. & Higgins, P. J. 1990. Handbook of Australian, New Zealand and Antarctic birds, vol. 1. 
Oxford Univ. Press. 

Martinez, J. 1987. Un nouveau cas probable d'endemisme insulaire: le canard de l'ile Amsterdam. 
Documents Lab. Geologie Faculte Sci. Lyon 99: 211-219. 

McCormick, R. 1842. A sketch of the Antarctic regions. Tasman. J. Nat. Sci. 1: 214-246. 

McCracken, K. G. & Sorenson, M. D. 2004. Is homoplasy or lineage sorting the source of incongruent 
mtDNA and nuclear gene trees in the stiff-tailed ducks (Nomonyx-Oxyura)? Systematic Biol. 54: 
35-55. 

Milberg, P. & Tyrberg, T 1993. Naive birds and noble savages — a review of man-caused prehistoric 
extinctions of island birds. Ecography 16: 229-250. 

Millener, P. R. 1996. Extinct birds. Pp. 113-120 in Department of Conservation The Chatham Islands: 
heritage and conservation. Canterbury Univ. Press, Christchurch. 

Millener, P. R. 1999. The history of the Chatham Islands' bird fauna of the last 7000 years — a chronicle 
of change and extinction. Smithsonian Contrib. Paleobiology 89: 85-109. 

Miller, G H., Fogel, M. L., Magee, J. W., Gagan, M. K., Clarke, S. J. & Johnson, B. J. 2005. Ecosystem 
collapse in Pleistocene Australia and a human role in megafaunal extinction. Science 309: 287-290. 

Mourer-Chauvire, C, Bour, R., Ribes, S. & Moutou, F. 1999. The avifauna of Reunion Island (Mascarene 
Islands) at the time of the arrival of the first Europeans. Pp. 1-38 in Olson, S. L. (ed.) Avian paleon- 
tology at the close of the 20th century: Proc. 4th Intern. Meeting Soc. Avian Paleontology and 
Evolution, Washington, D.C, 4-7 June 1996. Smithsonian Contrib. Paleobiology 89. Smithsonian 
Institute, Washington DC. 

Olson, S. L. & James, H. F. 1984. The role of Polynesians in the extinction of the avifauna of the Hawaiian 
Islands. Pp. 768-780 in Martin, P. S. & Klein, R. G (Eds.) Quaternary extinctions: a prehistoric rev- 
olution. Univ. of Arizona Press, Tucson. 

Pidgeon, M. 1996. An ecological survey of Lake Alaotra and selected wetlands of central and eastern 
Madagascar in analysing the demise of Madagascar pochard Aythya innotata. Unpubl. report. 

Salvadori, T 1894. Remarks on the ducks of the genera Anas and Nyroca. Bull. Brit. Orn. CI. 2: i-ii. 

Salvan, J. 1970. Remarques sur revolution de l'avifauna malgache depuis 1945. Alauda 38: 191-203. 

Skerrett, A. 1999. Ferruginous Duck Aythya nyroca: the first record for Seychelles. Bull. Afr. Bird CI. 6: 
148. 

Soothill, E. & Whitehead, P. 1978. Wildfowl of the world. Blandford, Poole. 

Sorenson, M. D. & Fleischer, R. C. 1996. Multiple independent transpositions of mitochondrial DNA con- 
trol region sequences to the nucleus. Proc. Nat. Acad. Sci. USA 93: 15239-15243. 

St John, J., Cotter, J. -P. & Quinn, T W. 2005. A recent chicken repeat 1 retrotransposition confirms the 
Coscoroba-Cape Barren goose clade. Mol. Phyl. & Evol. 37: 83-90. 

Steadman, D. W. & Martin, P. S. 2003. The late Quaternary extinction and future resurrection of birds on 
Pacific islands. Earth-Sciences Reviews 61: 133-147. 

Taylor, R. H. 1979. How the Macquarie Island Parakeet became extinct. New Zealand J. Ecol. 2: 42-45. 

Tennyson, A. J. D. & Millener, P. R. 1994. Bird extinctions and fossils from Mangere Island, Chatham 
Islands. Notornis (Suppl.) 41: 165-178. 



H. Glyn Young & Janet Kear 39 Bull. B.O.C. 2006 126A 

Turbott, E. G. (convenor). 1990. Checklist of the birds of New Zealand and the Ross Dependency, 

Antarctica. Third edn. Random Century, Auckland. 
Threatened Waterfowl Specialist Group (TWSG). 2003. Threatened waterfowl species and subspecies. 

Threatened Waterfowl Specialist Group News 14: 2-A. 
Vickers-Rich, P. & Rich, T H. 1 999. Wildlife of Gondwana: dinosaurs and other vertebrates from the 

ancient supercontinent. Indiana Univ. Press, Indianapolis. 
Webb, C. S. 1936. Collecting waterfowl in Madagascar. Avicult. Mag. (5)1: 36-39. 
Weller, M. W. 1980. The island waterfowl. Iowa State Univ. Press, Ames. 

Williams, M. 2003. Waterfowl of an isolated southern mini-continent: New Zealand. Casarca 9: 53-99. 
Wilme, L. 1993. A recent record of the rare Madagascar Pochard (Aythya innotata) on Lake Alaotra, 

Madagascar. Bull. Brit. Orn. CI. 113: 188-189. 
Wilme, L. 1994. Status, distribution and conservation of two Madagascar bird species endemic to Lake 

Alaotra: Delacour's Grebe Tachybaptus rufolavatus and Madagascar Pochard Aythya innotata. Biol. 

Consent. 69: 15-21. 
Worthy, T. H. & Holdaway, R. N. 2002. The lost world of the Moa: prehistoric life of New Zealand. Indiana 

Univ. Press, Bloomington. 
Worthy, T H., Holdaway, R. N., Sorenson, M. D. & Cooper, A. C. 1997. Description of the first complete 

skeleton of the extinct New Zealand goose Cnemiornis calcitrans (Aves: Anatidae), and a reassess- 
ment of the relationships of Cnemiornis . J. Zool, Lond. 243: 695-723. 
Young, H. G 1996. Threatened Anatinae and wetlands of Madagascar: a review and evaluation. In Birkan, 

M., van Vessem, J., Havet, P., Madsen, J., Trolliet, B. & Moser, M. (eds.) Proc. Anatidae 2000 Conf, 

Strasbourg, France, 5-9 December 1994. Gibier Faune Sauvage, Game & Wildlife 13: 801-813. 
Young, H. G 2002. Predicting the ecology of Madagascar's endemic dabbling ducks using captive popu- 
lations and related taxa: implications for conservation. Ph.D. thesis, University of Kent, Canterbury. 
Young, H. G. 2005. White-backed Duck Thalassornis leuconotus. Pp. 210-212 in Kear, J. (ed.) Ducks, 

geese and swans. Oxford Univ. Press. 
Young, H. G, Lewis, R. E. & Razafindrajao, F. 2003. Anseriformes: Anatidae, Wildfowl. Pp. 1071-1077 

in Goodman, S. M. & Benstead, J. P. (eds.) The natural history of Madagascar. Univ. of Chicago 

Press. 
Young, H. G & Rhymer, J. M. 1998. Meller's Duck: a threatened species receives recognition at last. 

Biodiver. & Conserv. 7: 1313-1323. 
Young, H. G & Smith, J. G 1989. The search for the Madagascar Pochard Aythya innotata: survey of Lac 

Alaotra, Madagascar October-November, 1989. Dodo, J. Jersey Wildlife Preserv. Trust 26: 17-34. 
Young, H. G, Sorenson, M. D. & Johnson, K. P. 1997. A description of the Madagascar Teal Anas bernieri 

and an examination of its relationships with the Grey Teal A. gracilis. Wildfowl 48: 174-180. 
Young, H. G, Tonge, S. J. & Hume, J. P. 1996. Review of Holocene wildfowl extinctions. Wildfowl 47: 

166-180. 

Addresses: H. Glyn Young, Durrell Wildlife Conservation Trust, Les Augres Manor, Trinity, Jersey JE3 
5BP, UK, e-mail: Glyn.Young@durrell.org. Janet Kear| 



( idle Mourer-Chauvire el al. 40 Bull. B.O.C. 2006 126 A 

Recent avian extinctions on Reunion 

(Mascarene Islands) from paleontological 

and historical sources 

by Cecile Mourer-Chauvire, Roger Bour & Sonia Ribes 

The Mascarene Islands were uninhabited when the first Europeans settled there, 
during the 16th century. The strange avifauna of these islands was described by the 
early travellers, but many species disappeared very rapidly. Fossil remains were 
discovered very early on Rodrigues, and later on Mauritius, but it was only in 1974 
that the first remains of fossil birds were discovered on Reunion, in a large cave, 
'Grotte des Premiers Francais'. Subsequently, other remains were discovered in 
small basaltic caves and in a marsh. These fossil birds were studied by Cowles 
(1987, 1994), Mourer-Chauvire & Moutou (1987), Mourer-Chauvire et al. (1994, 
1995a,b), following which a comprehensive paper concerning the different species 
and fossiliferous localities was issued (Mourer-Chauvire et al. 1999). 

The original avifauna of Reunion is also known from the accounts of early 
visitors, whose reports were collated by Lougnon (1970). The parts concerning birds 
are also presented by Barre & Barau (1982) and Barre et al. (1996). Particular parts 
of these accounts were discussed by Cheke (1987), and a previously unknown 
report, by Melet, who called at Reunion in 1671, was discovered by Anne Sauvaget 
and published in 1999 (Sauvaget 1999). 

The solitaire 

Several early travellers mentioned a large, almost flightless, whitish bird called the 
solitaire which, following the discovery of 17th-century paintings, was generally 
admitted to be a white dodo. However, in 1980 the remains of a large, extinct insular 
ibis, initially named Borbonibis latipes (Mourer-Chauvire & Moutou 1987), were 
discovered. It was some surprise that this bird had never been mentioned in the early 
reports. As further remains of the ibis were found in other localities in the west of 
the island, it must have been relatively abundant in areas near the coast, i.e. where 
the first navigators landed. During our excavations, and despite our expectations, we 
never found a fossil bone attributable to a dodo. Re-analysis of the early accounts 
led us to the conclusion that the Reunion solitaire was not a dodo but the above- 
mentioned ibis. In terms of its osteological characteristics, the ibis is related both to 
Threskiornis aethiopicus, Sacred Ibis, and T. spinicollis, Straw-necked Ibis. 

The descriptions of the solitaire, given by the travellers, agree well with an ibis 
and some cannot apply to a dodo. Abbe Carre, who was on Reunion in 1 667, stated: 
Tt would look like a turkey, if it did not have higher legs. The beauty of its plumage 
is a delight to see. It is of changeable colour which verges upon yellow'. Melet, who 
visited Reunion in 1671, wrote: 'and other kinds of birds that are called Solitaires, 
which are very tasty, and the beauty of their plumage is very curious by the diversity 



Cecile Mourer-Chauvire et al. 41 Bull. B. O. C. 2006 1 26A 

of bright colours which shone on their wings and around their neck'. Sieur Dubois, 
who was present on Reunion in 1671-72, elaborated: 'These birds are so-called 
(Solitaires) because they always go alone. They are as big as a large goose and have 
white plumage, black at the tip of the wings and tail. At the tail there are feathers 
approaching those of the ostrich. They have a long neck and the beak made like that 
of the woodcock, but bigger, and legs and feet like the turkeys. This bird is caught 
by running after it, since it flies only very little'. Feuilley, visiting Reunion in 1704, 
added: 'The Solitaires are the size of an average turkey cock, grey and white in 
colour. They inhabit the tops of the mountains. Their food is only worms and filth, 
taken on or in the soil' (after Barre & Barau 1982, Sauvaget 1999, our translation). 
The long legs, black wingtips and tail, long neck, beak like that of a woodcock but 
stronger and food consisting of worms taken on or in the soil do not correspond to 
a dodo, which had a strong, inflated bill and was frugivorous. The bright colours on 
the wings and neck are reminiscent of the iridescent plumage of Straw-necked Ibis. 
Sieur Dubois is the only author to mention that the solitaire was still able to fly. 

The only accounts contra this interpretation are those of Tatton and Bontekoe. 
Tatton, who was on Reunion in 1613, wrote: 'A great fowl, the bigness of a Turkie, 
very fat and so short-winged that they cannot flie, being white, and in a manner 
tame', single and Bontekoe, who called in at Reunion in 1619, suggested: 'There 
were also some Dod-eersen, which had small wings but could not fly; they were so 
fat that they could scarcely walk, for when they walked their belly dragged along 
the ground' (Fuller 2002). But Bontekoe's ship was blown up and his account was 
published only in 1646. It is likely that 'having a recollection of a large brevipennate 
bird in Bourbon, whose tameness rendered it an easy prey to his sailors, he 
concluded it to be a Dodo, and adopted the name and descriptions of that bird which 
had been given by previous navigators' (Strickland & Melville 1848). For their part, 
Hume & Cheke (2004) have demonstrated that all of the paintings of white dodos 
were based on an early picture, by Roelant Savery, of a whitish specimen of a 
Mauritius Dodo Raphus cucullatus, painted in Prague c.1611. 

Another objection to the occurrence of a true dodo on Reunion is the island's 
very recent geological age. Recent work on the DNA of Mauritius Dodo and 
Rodrigues Solitaire Pezophaps solitaria have revealed these two to be sister taxa 
and that their closest living relative is Nicobar Pigeon Caloenas nicobarica. The 
maximum likelihood molecular clock indicates 'that the dodo/solitaire and 
Caloenas diverged in the mid/late Eocene, around 42.6 million years ago (Ma) (95% 
confidence interval = 31.9 to 56.1 Ma), whereas the dodo and the solitaire separated 
in the late Oligocene, about 25.6 Ma (17.6 to 35.9 Ma)' (Shapiro et al. 2002). These 
analyses suggest that the lineages of the dodo and Rodrigues Solitaire diverged a 
very long time ago. The age of Mauritius and Rodrigues is estimated at 8-10 Ma 
(Hume & Cheke 2004), whilst Reunion only dates 3 Ma (Molnar & Stock 1987). 
Much of the Reunion fauna and flora derives from Mauritius and it is probable that, 
when Reunion emerged from the sea, the ancestors of the dodo and Rodrigues 



Cecile Mourer-Chauvire et al. 42 Bull. B.O.C. 2006 126A 

Solitaire had already lost their ability to fly, and were thus no longer able to colonise 
the newly appeared island (Hume & Cheke 2004). 

In conclusion we consider that, for now, there is no paleontological or pictorial 
evidence for the existence of a white dodo on Reunion, and that the descriptions of 
the solitaire agree better with the extinct ibis which has been found in several 
localities and is now named Threskiornis solitarius (Selys-Longchamps, 1848) 
(Mourer-Chauvire et al. 1995b). 

Other species, extinct or extirpated from Reunion, 
described by the navigators and found as fossils 

Nycticorax duboisi (Rothschild, 1907), Reunion Night Heron 

The description given by Dubois is: 'Bitterns or Great gullets, large as big capons 
{G alius gal I us) but fat and good (to eat). They have grey plumage, each feather 
tipped with white, the neck and beak like a heron and the feet green, like the feet of 
the 'Poullets dTnde' (Meleagris gallopavo). That lives on fish' (Barre & Barau 
1982, our translation). Two probably flightless species of night herons also formerly 
occurred in the Mascarenes, on Mauritius {Nycticorax mauritianus) and Rodrigues 
(N. megacephalus). Unlike these two forms, the proportions of the wing and leg 
bones of the Reunion heron show that it had flight capabilities quite similar to that 
of living species. 

Phoenicopterus ruber Linnaeus, 1758, Greater Flamingo 

Greater Flamingos were mentioned several times in historical accounts of Reunion 
(Lougnon 1970) and Feuilley indicated that there were 3,000^1,000 of them in 1704 
on the Etang du Gol (Barre & Barau 1982). They disappeared between 1710 and 
1730 (Cheke 1987). 

Alopochen (Mascarenachen) kervazoi (Cowles, 1994), Kervazo's Egyptian 
Goose 

Several reports mention the presence of geese, but only until the time of Dubois, 
who gave the most detailed description: 'wild geese, slightly smaller than the 
European geese. They have the same feathering, but with the bill and feet red. They 
are very good [to eat]' (Barre & Barau 1982, our translation). The identification of 
this goose as closely related to Alopochen aegyptiacus, Egyptian Goose, agrees well 
with Dubois' description of the red bill and the feet. The extinct goose of Reunion, 
and the extinct Malagasy Goose Alopochen sirabensis, exhibit a slight reduction in 
the length of the wing elements (ulna and carpometacarpus), and a slight increase in 
the length of the femur. 

Anas theodori Newton & Gadow, 1893, Sauzier's Teal, and cf. Aythya sp., a Pochard 

Dubois mentions: 'River ducks, smaller than European ones, feathered like teals. 
They arc good [to eat]' (Barre & Barau 1982, our translation). The Anas remains 



Cecile Mourer-Chauvire et al. 43 Bull. B.O.C. 2006 126A 

from Reunion are attributed to the same species as that described from Mauritius. 
The dimensions of the bones do not indicate a diminished flying ability. It is thus 
possible that the species could fly between Mauritius and Reunion. 

Falco duboisi Cowles, 1994, Reunion Kestrel 

Dubois and another author mentioned kestrels. Dubois noted three different birds of 
prey. The first were Reunion Harriers Circus maillardi, a still-extant species. 'The 
second ones are named yellow-feet, with the size and shape of falcons' (Barre & 
Barau 1982, our translation). The Reunion Kestrel is much larger than the endemic 
insular kestrels Falco araea (Seychelles Kestrel) and F. newtoni (Madagascar 
Kestrel), and slightly larger than F punctatus (Mauritius Kestrel). It also differs 
from the latter by its less-reduced wings. 

Dryolimnas augusti Mourer-Chauvire et al., 1999, Reunion Wood Rail 

Dubois was the only author to mention rails, and he simply wrote that there were 
wood rails. The rail remains have been attributed to the genus Dryolimnas and they 
differ from the recent species, D. cuvieri, White-throated Rail, in their larger size 
and by the shape of the tarsometatarsus which is much more robust. The proportions 
of the wing elements compared to those of the leg reveal that the Reunion Wood 
Rail was flightless, a characteristic of the extant subspecies D. cuvieri aldabranus, 
of Aldabra, whilst the nominate taxon from Madagascar remains volant. 

Fulica newtonii Milne-Edwards, 1867, Newton Coot 

A large, extinct species of coot was described by Milne-Edwards from remains 
found at the Mare aux Songes, Mauritius. The remains found on Reunion do not 
differ. This form is related to extant Fulica cristata, Red-knobbed Coot, which is 
principally found in Africa and Madagascar, but is slightly larger. The proportions 
of the bones indicate some reduction in flying ability. Many authors have reported 
the presence of water hens, but the most detailed description is given by Dubois: 
'Water hens, which are as big as hens. They are completely black and have a big 
white crest on the head' (Barre & Barau 1982, our translation). 

Nesoenas duboisi Rothschild, 1907, Reunion Pink Pigeon 

All the early navigators, including Feuilley in 1704, noted the abundance of pigeons 
and doves. Despite this, we have found only two remains attributable to this pigeon. 
They differ from the genus Alectroenas, being more similar to Mauritius Pink 
Pigeon Nesoenas mayeri, and correspond to the description given by Dubois of 
russet-red wild pigeons: 'They are a little larger than the European Pigeons, and 
have a stronger bill, red at the end close to the head, the eyes ringed by flame colour, 
like the pheasants' (Barre & Barau 1982, our translation). A molecular study of the 
genera Streptopelia and Columba has shown Mauritius Pink Pigeon to be sister to 
Streptopelia picturata, Madagascan Turtle Dove (Johnson et al. 2001). These 
authors suggested the genus Nesoenas be merged with Streptopelia, but Cheke 



Cecile Mourer-Chauvire et al. 44 Bull. B.O.C. 2006 126A 

(2005) emphasised the similarities between picturata and mayeri and their 
ditTerenees compared to other Streptopelia, and proposed retention of Nesoenas 
which should also include picturata (the position adopted here). 

Mascarinus mascarinus (Linnaeus, 1771), Mascarene Parrot 

The early travellers also spoke of the very large quantity and diversity of parrots, of 
which at least four species occurred on Reunion. We have found few remains 
attributable to the Mascarene Parrot. Dubois' description, 'Parrots a little bigger 
than pigeons, the feathering of the colour of petit-gris, a black hood on the head, the 
beak very strong and the colour of fire' (Barre & Barau 1982, our translation), 
agrees very well with Mascarinus mascarinus which is known from late- 18th- 
century illustrations. Petit-gris is the name given to the fur of Eurasian Red Squirrel 
Sciurus vulgaris in its dark phase. 

Fregilupus varius (Boddaert, 1783), Reunion Starling 

We have found one bone of Reunion Starling. The description given by Dubois 
agrees well with this species which is known from 18th-century illustrations and 
specimens: 'Hoopoes or 'Callendres', with a white tuft above the head, the rest of 
the plumage white and grey, the beak long, and the feet as that of a bird of prey' 
(Barre & Barau 1982, our translation). 

Species mentioned historically for which fossils and 
specimens are unavailable 

Cyanornis (l=Porphyrio) caerulescens (Selys-Longchamps, 1848), Oiseau bleu 
The presence of a large, blue bird was first mentioned by Dubois, without indication 
of locality, but all subsequent writers reported that it was found in the plains above 
the mountains and mainly at 'Plaine des Cafres', in the south-west part of the island, 
more than 1,500 m above sea level. As yet, no fossil remains have been found, 
which is probably attributable to its distribution; it was not present in that part of the 
island where all the fossiliferous localities are located. 

Dubois wrote: 'Oiseaux bleus, as big as solitaires. Their plumage is all blue, 
their bill and feet red, made as the feet of fowls. They do not fly but they run 
extremely fast, so that a dog has difficulty to catching them when hunting. They are 
very good [to eat]' (Barre & Barau 1982, our translation). According to a report 
attributed to a certain Father Brown (see Cheke 1987): 'It rarely flies, always 
hugging the ground, but it runs with surprising speed' (Olson 1977). Taking into 
account its size and that it was not entirely flightless, Olson (1977) and Cheke 
(1987) considered it possible that this bird belonged to the genus Porphyrio. 



Cecile Mourer-Chauvire et al. 45 Bull. B.O.C. 2006 126A 

Extinct species not reported by the early travellers 

Mascarenotus grucheti Mourer-Chauvire et al, 1994, Gruchet's Lizard Owl 

We found, in several localities, the remains of a Strigid owl never reported in the 
historical accounts. We referred it to a new genus, Mascarenotus, which includes the 
three Mascarene owls, M. murivorus (Milne-Edwards, 1873), of Rodrigues, M. 
sauzieri (Newton & Gadow, 1893), of Mauritius, andM grucheti, of Reunion. The 
last two have a much-elongated tarsometatarsus, a characteristic of other, extant 
insular species, e.g. Gymnoglaux lawrencii, Cuban Screech-owl, Otus nudipes, 
Puerto Rican Screech-owl, and in the extinct insular species of the genus Grallistrix, 
from Hawaii (Olson & James 1991). The tarsometatarsi of M. grucheti were almost 
the same size as those of M. sauzieri, whilst the wing bones, of which only an 
incomplete humerus is available, were slightly smaller than that of M. sauzieri. 
However, two left ulnae have recently been found in a new locality, the Caverne 
Payet, at Grande Chaloupe, on the north coast of the island. These confirm that in 
Gruchet's Lizard Owl the wings were more reduced than in the other Mascarene 
owls (Fig. 1, Table 1). 



2 cm 

1 





B 



Figure 1 . Two left ulnae of Mascarenotus grucheti from Caverne Payet, Grande Chaloupe, Reunion (B 
and C), Museum d'Histoire naturelle de La Reunion, nos. MHN-RUN-CP-0 1966-67, compared with an 
ulna of Mascarenotus sauzieri, from Montagne du Pouce, Mauritius (A), Museum national d'Histoire 
naturelle, Paris, no. MAD 7192. Natural size. 



Cecile Mourer-Chauvire el al. 



46 



Bull. B.O.C. 2006 126 A 



TABLE 1 

Measurements of the humeri and ulnae of the three species of Mascarenotus. 

{ 1 ) materia] from Marc aux Songes and Montague du Pouce (Thirioux collection), Mauritius, Museum 

national d'Histoire naturelle. Paris. (2) after Giinther & Newton 1879. (3) humerus from Grotte des 

Premiers Francais, do. LAC 1993-50. MNHN Paris, and ulnae from Caverne Payet, nos. MHN-RUN- 

CP-0 1966-67, Museum d'Histoire naturelle de La Reunion. 



Humerus 


Ulna 






\/. sauzieri, Mauritius 








Total length ( 1 ) 


extremes 


65.0-71.5 


72.0-86.0 




mean (/?) 


67.10(5) 


79.25 (6) 


\/. murivorus, Rodrigues 








Total length (2) 


extremes 


64.0-69.0 


- 




mean (/?) 


67.50 (2) 


74.0(1) 


\l. grucheti, Reunion 








Total length (3) 


extremes 


- 


64.0-66.4 




mean (/?) 


est. 60.5(1) 


65.20 (2) 



Species extant on Reunion also known from fossils 

Puffinus paciflcus (Gmelin, 1789), Wedge-tailed Shearwater 

Puffinus Iherminieri Lesson, 1840, Audubon's Shearwater 

Phaethon lepturus Daudin, 1 802, White-tailed Tropicbird 

Nwnenius phaeopus (Linnaeus, 1758), Whimbrel 

Nesoenas picturata (Temminck, 1813), Madagascar Turtle Dove (Cheke 2005) 



Chronology of the extinctions 

From the early accounts and the fossil record, at the time when Europeans settled 
Reunion, the avifauna included at least 33 species of resident landbirds. Of these, 17 
are extinct, five no longer occur on Reunion, and 1 1 are extant there. Amongst the 
1 1 survivors, eight are very small (Apodiformes and Passeriformes) and Coracina 
newtoni (Pollen), Reunion Cuckoo-shrike, is globally threatened. 

The historical reports enable us to follow the chronology of extinctions, which 
occurred very rapidly, over a period of two centuries from 1646. A first group of 
species, reported by the earliest visitors and by Dubois in 1671-72, apparently 
became extinct almost immediately because they were not reported thereafter, 
namely: Nycticorax duboisi, Alopochen (M) kervazoi, Anas theodori, a pochard, 
Falco duboisi, possibly a smaller falcon known as Emerillon, Dryolimnas augusti, 
Fulica newtonii, a parrot known as 'Perroquet vert a tete... couleur de feu', and a 
Foudia sp. Rats were absent in 1671, as indicated in the log of Le Breton and by 
Dubois, but had invaded by 1675 (Cheke 1987). The first wave of extinction mainly 
included aquatic forms inhabiting the ponds and marshes of the west coast, the first 
area to be settled. 



Cecile Mourer-Chauvire et al. 47 Bull. B.O.C. 2006 126A 

The second wave of losses included species mentioned by Feuilley in 1 704 but 
not recorded thereafter. These included a cormorant, probably Phalacrocorax 
africanus (Long-tailed Cormorant), an egret, probably Egretta garzetta dimorpha 
(Dimorphic Egret), Phoenicopterus ruber, Alectroenas sp., a blue -pigeon, Nesoenas 
duboisi, and perhaps another dove. The Reunion solitaire Threskiornis solitarius 
survived for a short time, taking refuge in the mountains, but was reported for the 
last time in 1708. Cats were introduced in 1703 to exterminate rats and must have 
played a significant role in the destruction of the birds. Thereafter, in 1734-63, 
Oiseau bleu (Cheke 1987), a grey parrot and a parakeet, Psittacula equeslecho, 
disappeared, followed c.1780 by Mas carinus mascarinus, and lastly, in 1838-58, by 
Fregilupus varius (Barre & Barau 1982). 

It is well known that Reunion's birds were so tame that a single man, armed with 
a simple stick, could kill up to 200 of them in one day (see Mourer-Chauvire et al. 
1999). Man has certainly played a direct role in the extinction of these birds, and 
that of the giant land tortoise, Cylindraspis indica (Austin et al. 2002), but has also 
played an indirect role by introducing predators (pigs, rats, cats) or competitors, and 
by destroying habitats. Finally, it is also possible that the introduction of a disease 
or parasite wrought the extinction of the Reunion Starling (Cheke 1987). 

References: 

Austin, J. J., Arnold, E. N. & Bour, R. 2002. The provenance of type specimens of extinct Mascarene 

Island giant tortoises {Cylindraspis) revealed by ancient mitochondrial DNA sequences. J. 

Herpetology 36: 280-285. 
Barre, N. & Barau, A. 1982. Oiseaux de la Reunion. Imprimerie Arts graphiques moderaes, La Reunion. 
Barre, N., Barau, A. & Jouanin, C. 1996. Oiseaux de la Reunion. Les Editions du Pacifique, Paris. 
Cheke, A. S. 1987. An ecological history of the Mascarene Islands, with particular reference to extinc- 
tions and introduction of land vertebrates. Pp. 5-89 in Diamond, A. W. (ed.) Studies of Mascarene 

Islands birds. Cambridge Univ. Press. 
Cheke, A. S. 2005. Naming segregates from the Columba-Streptopelia pigeons following DNA studies 

on phylogeny. Bull. Brit. Orn. CI. 125: 293-295. 
Cowles, G. S. 1987. The fossil record. Pp. 90-100 in Diamond, A. W. (ed.) Studies of Mascarene Islands 

birds. Cambridge Univ. Press. 
Cowles, G. S. 1994. Anew genus, three new species and two new records of extinct holocene birds from 

Reunion Island, Indian Ocean. Geobios 27: 87-93. 
Fuller, E. 2002. Dodo, from extinction to icon. HarperCollins, London. 

Hume, J. P. & Cheke, A. S. 2004. The White Dodo from Reunion Island: unravelling a scientific and his- 
torical myth. Archiv. Nat. Hist. 31: 57-79. 
Johnson, K. P., Kort, S. de, Dinwoodey, K., Mateman, A. C, Cate, C. ten, Lessels, C. M. & Clayton, D. 

H. 2001. A molecular phylogeny of the dove genera Streptopelia and Columba. Auk 118: 874-887. 
Giinther, A. & Newton, E. 1879. The extinct birds of Rodriguez. Phil. Trans. Roy. Soc, Lond. 168: 

423-427. 
Lougnon, A. 1970. Sous le signe de la tortue. Voyages anciens a Vile Bourbon (1611-1725). Third edn. 

Privately published, Saint-Denis, Reunion. [A fourth edition, the text identical to that of 1970, was 

published in 1992 by Azalees Editions.] 
Molnar, P. & Stock, J. 1987. Relative motions of hotspots in the Pacific, Atlantic and Indian Oceans since 

Late Cretaceous Time. Nature 327: 587-591. 
Mourer-Chauvire, C, Bour, R. & Ribes, S. 1995a. Was the solitaire of Reunion an ibis? Nature 373: 568. 



Cecile Mourer-Chauvire et al. 48 Bull. B.O.C. 2006 126A 

Mourer-C'ham ire. C, Bour, R. & Ribes, S. 1995b. Position systematique du Solitaire de la Reunion: nou- 
velle interpretation basee sur les restes fossiles et les recits des anciens voyageurs. C. R. Acad. Sci. 
Paris 320 serie Ha: 1125 1131. 

Mourer-Chauvire. C, Bour. R.. Moutou. F. & Ribes, S. 1994. Mascarenotus nov. gen. (Aves, 
Strigi formes), genre endemique eteint des Mascareignes et M. grucheti n. sp., espece eteinte de La 
Reunion. C R. Acad. Sci. Paris 318 serie II: 1699-1706. 

Mourer-Chauvire, C. Bour, R., Ribes, S. & Moutou, F. 1999. The avifauna of Reunion Island (Mascarene 
Islands) at the time of the arrival of the first Europeans. Pp. 1-38 in Olson, S. L. (ed.) Avian paleon- 
tology at the close of the 20th century: Proc. 4th Intern. Meeting Soc. Avian Paleontology and 
Evolution. Washington, D.C., 4-7 June 1996. Smithsonian Contrib. Paleobiology 89. Smithsonian 
Institute. Washington DC. 

Mourer-Chauvire, C. & Moutou, F. 1987. Decouverte d'une forme recemment eteinte d'ibis endemique 
insulaire de Tile de la Reunion : Borbonibis latipes n. gen. n. sp. C. R. Acad. Sci. Paris 305 serie II: 
419—423. 

Olson. S. L. 1977. A synopsis of the fossil Rallidae. Pp. 339-379 in Dillon Ripley, S. (ed.) Rails of the 
world. David R. Godine, Boston. 

Olson, S. L. & James, H. F. 1991. Descriptions of thirty-two new species of birds from the Hawaiian 
Islands: Part I. Non-Passeriformes. Orn. Monogr. 45: 1-88. 

Sauvaget, A. 1999. La relation de Melet du voyage de La Haye aux Indes orientales. Etudes Ocean Indien 
25-26 (1998): 95-289. 

Shapiro, B., Sibthorpe, D., Rambaut, A., Austin, J., Wragg, G. M., Bininda-Edmonds, O. R. P., Lee, P. L. 
M. & Cooper, A. 2002. Flight of the dodo. Science 295: 1683. 

Strickland, H. E. & Melville, A. G 1848. The Dodo and its kindred, or the history, affinities and osteol- 
ogy of the Dodo, Solitaire, and other extinct birds of the islands Mauritius, Rodriguez, and Bourbon. 
Reeve, Benham & Reeve, London. 

Addresses: Cecile Mourer-Chauvire, UMR 5125, Paleoenvironnements et Paleobiosphere, Centre des 
Sciences de la Terre, Universite Claude Bernard — Lyon 1, 27^43 Boulevard du 1 1 Novembre, 69622 
Villeurbanne Cedex, France, e-mail: Cecile.Mourer@univ-lyonl.fr. Roger Bour, Laboratoire des 
Reptiles et des Amphibiens, Museum national d'Histoire naturelle, 25 rue Cuvier, 75005 Paris, 
France. Sonia Ribes, Museum d'Histoire Naturelle, 1 rue Poivre, 97400 Saint-Denis, La Reunion, 
France. 



Julian Pender Hume et al. 49 Bull. B.O.C. 2006 126A 

Unpublished drawings of the Dodo Raphus 
cucullatus and notes on Dodo skin relics 

by Julian Pender Hume, Anna Datta & David M. Martill 

The Dodo Raphus cucullatus was an endemic giant flightless pigeon from Mauritius 
that died out within 100 years of its discovery in 1598 (Moree 1998, Hume et al. 
2004) It has become a metaphor for extinction, exemplifying man's destructive 
capabilities on endemic oceanic island species (Fuller 2002). Our scant knowledge 
of the Dodo's morphology and autecology is derived largely from historical 
accounts, including contemporary paintings and ships' records, although there has 
been debate as to their scientific accuracy (Kitchener 1993). Knowledge of the 
skeletal anatomy of the Dodo is more detailed, being derived mainly from fossil 
remains discovered in the Mare aux Songes in the 1860s (Owen 1866). Very few 
Dodo remains reached European shores, and thus very few scientists have ever had 
'hands-on' experience of this enigmatic bird. Such was the paucity of tangible 
evidence for the existence of the Dodo that in the early 19th century many 
considered the species to have been mythical (Strickland & Melville 1848). Here we 
announce the discovery of 1 9th-century illustrations of a Dodo foot, executed by 
John Edward Gray, while searching the archives in the general library of the Natural 
History Museum, London. 

Although a number of exotic species were brought back to Europe in historic 
times, the inability to keep animals alive, or to preserve dead material on long sea 
voyages in the 1 600s, resulted in comparatively few zoological specimens reaching 
European shores. Despite suggestions to the contrary (e.g. Hachisuka 1953), as few 
as four or five Dodo specimens — maybe even fewer — reached Europe, and only one 
perhaps two birds arrived alive (Hume in press). Amongst the imported birds was 
the so-called 'Oxford Dodo', a specimen which today comprises the only extant 
skin remains. It has been suggested that the Oxford example is the same Dodo as 
that seen alive in London in 1638 (Strickland & Melville 1848), but no substantive 
evidence to support this claim exists. Further examples of soft tissue dodo 
specimens once existed in Copenhagen (head) and Prague (beak and foot), but today 
only the bones are preserved and their histories are uncertain. Furthermore, at least 
one other specimen of a dodo, if indeed it was actually so, was reported to have been 
deposited at the Anatomy School, Oxford (e.g. Newton & Gadow 1 896), but again, 
its provenance and subsequent history are unknown. 

Brief historical review 

The Oxford Dodo has a complex history, having been exhibited as a stuffed bird in 
the collection of horticulturist John Tradescant (Tradescant 1656), in 1656, and 
bequeathed to Elias Ashmole in 1659 (Strickland & Melville 1848). The specimen 
remained in the Ashmolean Museum until its transfer to the Oxford University 



Julian Pender Hume el al. 

■MB 



50 



Bull. B.O.C. 2006 126A 




Figure 1 . Newly discovered unsigned illustrations of the Dodo Raphus cucullatus head in dorsal and 
lateral views, executed by John Edward Gray, c. 1 824. 




Museum during the 1850s. There was a 
long-held belief that this, by then unique, 
stuffed Dodo was thrown onto a fire in 
1755, and that only the head and a foot 
were rescued from the flames (e.g. 
Strickland & Melville 1848, Fuller 
2002). In fact, its removal from 
exhibition was a curatorial decision 
made to preserve what was left of the by 
then highly degraded specimen (Ovenell 
1992). The salvaged remains included 
the skin of the head, some feathers and a 
foot. Today, all that remains of this 
specimen are two halves of the skin of 
the head, now with very few feathers, the 
skull, and the bones of the right foot with 
some scraps of skin and sinew (Figs. 
4-5). 

Figure 2. Only known illustration of the 'Oxford 
Dodo' foot alongside the 'London foot'. The 
Oxford foot is more gracile and 1 1% smaller than 
the latter. They are here interpreted as male 
(London) and female (Oxford). Annotations in 
Gray's hand give dimensions of the feet. 



Julian Pender Hume et al. 



51 



Bull B.O.C. 2006 126A 




Figure 3. Head prior to dissection, executed by William Clift. 




Another Dodo foot termed the 'London foot', which 
could be seen in a residence formerly called the Music 
House, situated near the West End of St Paul's church, 
London, was collected by Hubert alias Forges (Forges 
1665). It was presented to the Royal Society of London 
and transferred to the former British Museum, where it 
was exhibited along with the most famous Dodo painting 
(Strickland & Melville 1848), once owned by George 
Edwards and affectionately known as 'George Edward's 
Dodo', painted by Roelandt Savery in c.1626 (still held 
in the library of the Natural History Museum [NHM]). 
The last definite mention of this specimen including the 
soft tissue was c.1848 (e.g. Richardson 1851). The foot 
was mentioned again by Newton & Gadow (1896) as 
'still reposing in the British Museum, but without its 
integuments'. This suggests that like the Oxford 
specimen, the London specimen's soft tissue had decayed 
or been dissected and in fact the foot, as originally 
depicted in Strickland & Melville (1848), no longer 

Figure 4. Oxford foot bones. 



Julian Pender Hume el al. 



52 



Bull. B.O.C. 2006 126A 



existed. Therefore it is likely that today the so-called missing foot (e.g. Fuller 2002) 
consists only of bone (after being cast) and researchers looking for the soft tissue 
specimen are, in fact, searching for the wrong type of material. Thus, by the end of 
the 1800s very little tangible non-fossil Dodo material was available for study. 

The Oxford Dodo head was dissected and illustrated in 1847, along with the 
London foot (Strickland & Melville 1848). The Oxford foot was also dissected, but 
by this time it lacked most of its soft tissues and, until recently, was never thought 
to have been illustrated with integuments. 

Newly discovered illustrations 

During a search of the zoological drawings held at the NHM, London, one of us 
(AD) discovered a folder entitled 'Didus' (Linnaeus's second but junior synonym 
for the dodo) compiled by John Edward Gray (1800-75). Gray joined the staff of 
the then British Museum (now NHM) as an assistant in 1 824, becoming Keeper of 
Zoology in 1840 until his retirement in 1874 (Anon. 1904). Gray amassed a large 
collection of published natural history illustrations in scrapbooks and also produced 
some drawings of his own. The 'Didus' folder contained one double-sided sheet 
measuring 340 x 210 mm with illustrations in black ink on paper bearing an 1824 
watermark (Figs. 1-2). Gray presented these dodo illustrations to the Zoological 
Club of the Linnean Society on 24 April 1828 (Anon. 1828) and, therefore, the 
pictures must have been executed during this four-year period. A short note was 
published and this is the only mention made of Gray's dodo sketches we have 
managed to trace: 

'At the request, of the Chairman, Mr. Gray exhibited a sketch of the foot of the 
dodo, Didus ineptus, L., [Raphus cucullatus] preserved in the British Museum, and 
another sketch of that contained in the Ashmolean Museum of Oxford, and also a 
head remaining in the latter collection. He remarked that the feet agreed so perfectly 




Figure 5. Dissected Oxford head. 



Julian Pender Hume et al. 53 Bull. B. O. C. 2006 1 26A 

in characters as to leave no doubt of their having belonged to the same species, but 
that although they were of opposite sides, the one being left and the other right, they 
must have been obtained from different individuals, the Oxford specimen being one 
inch shorter than that of the British Museum. ' 

On one side of the drawings is illustrated the Oxford head in dorsal and lateral 
views (Fig. 1) whilst the other side illustrates, uniquely, the Oxford and London 
dodo feet (Fig. 2), with accompanying annotations including measurements. On the 
first sheet accompanying the dodo feet the following measurements are presented: 

Oxford Specimen. Length a. Right foot Length. 8 inches & half from joint to 
end of middle toe Museum [London] Specimen. B. left foot Length. 9 inch 
and a half 

On the second sheet accompanying the dodo head drawings the following notes 
are made: [dorsal view, left] 4 inches [across head], 2.1/4 [in front of eyes], 1.1/4 
[across tip of bill], [lateral view, right]; nakedish with scattered hairs ending in two 
or three heads [written on head]; cere naked hard skin (in the middle]; cover of this 
part is thin, horny, the bone solid, porous [on bill tip] 

Whilst examining Sir Richard Owen's correspondence in the same library, JPH 
found a hitherto unpublished illustration of a Dodo head. The watercolour is signed 
'WC (William Clift, 1775-1849, conservator of the Hunterian collection, London) 
and comprises an illustration of the head of the Oxford Dodo specimen prior to its 
dissection (Fig. 3). Of particular note in this illustration is the presence of many 
head feathers that have subsequently disappeared. The discovery of Gray's 
previously unpublished illustrations constitutes the only scientific documentation of 
all known skin specimens of the Dodo illustrated together. This is particularly 
important for comparative study. 

Discussion 

Based on the handwritten measurements by Gray, the Oxford right foot is c. 1 1 % 
smaller and more gracile than the London left foot, yet the tarsometatarsus bone of 
the Oxford foot has fully-fused epiphyses, indicating the animal to be adult (Fig. 4). 
Such a size discrepancy in a Columbiform has been interpreted as representing 
sexual dimorphism (Livezey 1993). Gray's illustration certainly indicates that the 
London foot is larger than the Oxford foot, but virtually nothing is known of dodo 
ecology. Therefore, any interpretations based on these drawings must be made 
cautiously. 

Acknowledgements 

Robert Prys-Jones (NHM, Tring), staff of the Photographic Library and General Library at NHM, 
London; Sandra Chapman of the Palaeontology Department, NHM; Andrew Kitchener (Edinburgh); and 
Ray Symonds (Cambridge) supplied data; and Darren Naish, Anthony Cheke and Errol Fuller supplied 
constructive comment. Staff at the Mauritius Institute, Port Louis, offered access to material in their care. 



Julian Pender Hume et al. 54 Bull. B.O.C. 2006 126A 

References 

Anon. 1828. Proceedings of Learned Societies on subjects connected with Zoology. Zoological Club of 

the Linnean Society. Zoological J. 3: 605. 
Anon. 1904. The history of the collections contained in the Natural History Departments of the British 

Museum, vol. 1. Trustees of the Brit. Mus., London. 
Fuller. E. 2002. Dodo— from extinction to icon. HarperCollins, London. 
Hachisuka, M. 1953. The dodo and kindred birds, or the extinct birds of the Mascarene Islands. H. F. & 

G. Witherby, London. 
Hubert alias Forges, R. 1665. A catalogue of many natural rarities with great industry, cost and thirty 

years travel in foreign countries collected by Robert Hubert alias Forges, Gent., and sworn servant 

to his Majesty. And daily to be seen at the place formerly called the Music House near the west end 

of St. Paul's church. London. 
Hume, J. P., Martill, D. M. & Dewdney, C. 2004. Dutch diaries and the dodo's demise. Nature 429: 622. 
Hume, J. P. (in press) The history of the dodo Raphus cucullatus and the penguin of Mauritius. J. Nat. 

Hist. (Colin Harrison Memorial Volume). 
Kitchener, A. 1993. On the external appearance of the dodo. Arch. Nat. Hist. 20: 279-301. 
Livezey, B. 1993. An ecomorphological review of the dodo {Raphus cucullatus) and solitaire (Pezophaps 

solitaria), flightless Columbiformes of the Mascarene Islands. J. Zool. 230: 247-292. 
Moree, P. J. 1998. A concise history of Dutch Mauritius. Kegan Paul International & International 

Institute of African Studies, London & Leiden. 
Newton, A. & Gadow, H. 1896. A dictionary of birds. London: A. & C. Black. 
Ovenell, R. F. 1992. The Tradescant dodo. Arch. Nat. Hist. 19: 145-152. 
Owen, R. 1866. Memoir on the dodo (Didus ineptus, Linn.). Taylor & Francis, London. 
Richardson, G. F. 1851. An introduction to geology, and its associate sciences, mineralogy, fossil botany, 

and palaeontology. H. G. Bohn, London. 
Strickland, H. E. & Melville, A. G 1848. The dodo and its kindred. Reeve, Benham & Reeve, London. 
Tradescant, J. 1656. Musaeum Tradescantianum. London. 

Addresses: Julian Hume (author for correspondence), Paleobiology Research Group, School of Earth & 
Environmental Sciences, University of Portsmouth, Portsmouth POl 3QL, UK; and The Bird Group, 
Department of Zoology, The Natural History Museum, Akeman Street, Tring, Herts. HP23 6AP, UK, 
e-mail: J.Hume@bun.com. Anna Datta, Library & Information Services, The Natural History 
Museum, Cromwell Road, London SW7 5BD, UK, e-mail: a.datta@nhm.ac.uk. David M. Martill, 
Paleobiology Research Group, School of Earth & Environmental Sciences, University of 
Portsmouth, Portsmouth POl 3QL, UK, e-mail: David.martill@port.ac.uk 



David L. Roberts & Anna Saltmarsh 55 Bull. B. O. C. 2006 1 26A 

How confident are we that a species is extinct? 

Quantitative inference of extinction 

from biological records 

David L. Roberts & Anna Saltmarsh 

In most cases, the extinction of a species is not directly observed and must instead 
be inferred from the record of sightings or collections of individual organisms. It is 
common in such cases to date extinction to the time of the last sighting. However, 
when a species becomes rare prior to extinction, it may exist for many years without 
detection and the time of last sighting can be a poor estimate of the time of 
extinction (Roberts & Solow 2003). Until recently, a species was regarded as extinct 
if it had not been observed for 50 years (Reed 1996). However, the usefulness of this 
criterion is dependent on the life-history characteristics of the species in question. A 
revision of the IUCN Red List categories resulted in a species being classified as 
Extinct only when exhaustive surveys failed to produce any observations over a 
time period appropriate to the species' life history and throughout its known 
historical range (IUCN 2001). However, such quantitative assessments of trends in 
range and abundance are costly, requiring extensive field studies over a long period 
of time (Burgman et al. 2000). 

Two types of sightings data may be available, (a) field observations and (b) data 
available for specimens, of which there are estimated to be c. 2. 5 billion specimens 
in biological collections (Suarez & Tsutsui 2004). Unlike in situ sightings, these 
records represent primary verifiable observations and are built directly on current 
taxonomic expertise. Such collections represent, for the vast majority of species, our 
only knowledge. If we cannot successfully monitor populations for the purpose of 
conservation assessments, it becomes almost impossible to predict their decline or 
extinction with any certainty (Roberts & Kitchener 2006). Several methods have 
been presented which provide a probabilistic basis for the extinction hypothesis 
using these sighting dates to construct a binary time series record (Solow 1993a,b, 
Burgman 1995, McCarthy 1998, Solow & Roberts 2003, Mclnerny et al. 2006). 
Essentially these methods provide the probability that another collection will be 
made given the characteristics of a time series. Recently, a statistical method 
(Roberts & Solow 2003), optimal linear estimation, has been used to estimate the 
actual extinction date. 

An illustrated example 

Ivory-billed Woodpecker Campephilus principalis was once widespread across the 
south-east United States (nominate principalis) with another race in Cuba (C p. 
bairdii) (Short 1982, Collar et al. 1992, 1994, Fuller 2001). However, between the 
19th century and 1930s it experienced a dramatic decline due to forest clearance 
(Fuller 2001). The species' requirement for large tracts of virgin forest was partially 



David L. Roberts & Anna Salimarsh 56 Bull. B.O.C. 2006 126A 

its downfall, as it had large territory requirements (Collar et al. 1994) with a 
maximum abundance of one pair per 16 km 2 (Tanner 1942, King 1978-79), with 
further data suggesting that the density was even lower (Collar et al. 1992). In 
addition, commercial collection may have played a part in the decline of the species 
(Collar et al. 1992), along with competitive interaction with Pileated Woodpecker 
Dryocopus pileatus (Short 1982). Much of our knowledge stems from the survey of 
James T. Tanner in the 1930s, when he found only a few individuals (Tanner 1942). 
At the time, hope for the survival of the species centred on the 80,000-acre Singer 
Tract in Louisiana (Collar et al. 1992). However, the last sighting in this area was 
in 1944 (Fuller 2001), and the population probably disappeared by 1948 when the 
remaining 3 1 1 km 2 were cleared (King 1978-79). The last authentic sighting in the 
United States of Ivory-billed Woodpecker occurred in the Apalachicola Swamp of 
Florida in 1952 (Fuller 2001). Many now regard the species as probably extinct 
(Short 1 982, Collar et al. 1 992, 1 994). 

Based on ten sightings in 1928-52 (1928, 1929, 1935, 1936, 1937, 1938, 1939, 
1941, 1944, 1952) (E. Fuller pers. comm.) we can estimate the extinction date using 
the method described by Roberts & Solow (2003) to 6 = 1969, eight years after the 
last sighting. The approximate 0.95 confidence interval for 6 is (1958, 1991). The 
width of this confidence interval is a result of the low sighting rate at the end of its 
sighting record. 

Reports still occur, but no verifiable evidence has been produced. It is thought 
that many of these sighting may be of the similar, but smaller, Pileated Woodpecker 
(Fuller 2001). Although the two species are easily distinguished, it is human nature 
to want to see the rarer of the two. In the case of the Thylacine, recent sightings have 
largely been disproved, in fact the distance at which an encounter is made can be 
doubled and the duration of the sighting halved (Anon. pers. comm.). 

Discussion 

Application of statistical methods for inferring extinction from sightings records 
and museum specimens will aid our understanding of the probability of whether a 
taxon has become extinct. However, inference of extinction based on sightings is 
difficult, largely due to the inference that can be drawn from a sighting record that 
is dependent on how the sighting rate varies. As with other types of ecological data, 
interpretation of sightings data requires an understanding of the underlying 
processes (Solow & Roberts 2003). 

Postscript: Avian resurrection 

It is almost impossible to determine with certainty whether a species is extinct. The 
apparent rediscovery of Ivory-billed Woodpecker in 2004 is certainly remarkable, as 
it had been consigned to the list of North America's Extinct or 'Probably Extinct' 
bird species (Fitzpatrick et al. 2005, 2006), although the rediscovery has been 
thrown into question (Nemesio & Rodrigues 2005, Jackson 2006, Sibley et al. 



David L. Roberts & Anna Saltmarsh 57 Bull. B. O. C. 2006 126A 

2006). However its rediscovery, after more than 50 years, begs the question how do 
we know when a species is extinct? (Roberts in press) 

We described above how a sighting record may be used to give a probabilistic 
basis to an extinction statement. However, we used the A=10 most recent sightings 
of the Ivory-billed Woodpecker to determine a possible extinction date. Using the 
method described by Roberts & Solow (2003), we estimated the extinction date to 
be $ = 1960, with an approximate 0.95 confidence interval for 6 is (1958, 1991). 
Although not significant in the traditional sense (95%), the extinction statement 
would still warrant further investigation. However, the model used assumes that the 
sightings are in the tail of the record, i.e. towards the end of the decline (Coles 
2001). In this case they are not; one only needs to examine the high sighting rate 
particularly during the 1930s. If the &=5 most recent sightings prior to its 
rediscovery in 2004 are used (1938, 1939, 1941, 1944, 1952) the estimated 
extinction date is 6 = 1969, with an approximate 0.95 confidence interval for 6 is 
(1958,2156). 

The Roberts & Solow (2003) method was adapted to calculate the significance 
level (or P-value) in testing the extinction hypothesis as the upper bound of an 
approximate 1 — a confidence interval for an unknown time of extinction, T E (Solow 
2005). Roberts (in press) used this to examine the extinction hypothesis for the 
Ivory-billed Woodpecker in 2004 based on the k=5 most recent sightings (1938, 
1939, 1941, 1944, 1952). Further, if the Solow & Roberts (2003) non-parametric 
test is used, both generate probability values greater than 0.05 (0.186 and 0.133 
respectively) and thus infers that the species should not have been considered 
extinct. If we take the last sighting to be 1944 as others have suggested (J. Jackson 
pers. comm.), then the significance levels are 0.056 (Roberts in press) and 0.047 
respectively. Although the latter may be considered not significant in the traditional 
sense, it would warrant further investigation. 

Finally, methods that are based on optimal linear estimation are a good choice 
for such analysis (Solow 2005). However, the choice of size of h, as seen here, is 
problematic, too large and it violates the assumption of extreme order statistics, too 
small and the power is low (Coles 2001). Based on experience, Solow (2005) 
suggested that when k is at least 5, the method works well. 

Acknowledgements 

We thank Errol Fuller for providing data on the sightings of Ivory-billed Woodpecker and Andy Solow 
for discussions on extinction modelling. 

References: 

Burgman, M. A., Grimson, R. C. & Ferson, S. 1995. Inferring threat from scientific collections. Conserv. 
Biol. 9: 923-928. 

Burgman, M., Maslin, B. R. Andrewartha, D., Keatley, M. R., Boek, C. & McCarthy, M. 2000. Inferring 
threat from scientific collections: power tests and an application to Western Australian Acacia 
species. Pp. 7-26 in Ferson, S. & Burgman, M. (eds.) Quantitative methods for conservation biolo- 
gy. Springer- Verlag, New York. 

Coles, S. 2001. An introduction to statistical modelling of extreme values. Springer- Verlag, London. 



David L. Roberts & Anna Saltmarsh 58 Bull. B.O.C. 2006 126A 

Collar, N. J.. Crosby. M. J. & Stattersfield, A. J. 1994. Birds to watch 2: the world list of threatened birds. 
BirdLife International. Cambridge, UK. 

Collar. N. J.. Gonzaga. L. P.. Krabbe, N., Madrono Nieto, A., Naranjo, L. G, Parker, T. A & Wege, D. C. 
1992. Threatened birds of the Americas: the ICBP/IUCN Red Data Book. International Council for 
Bird Preservation, Cambridge. UK. 

Fitzpatrick. J. W., Lammertink, M., Luneau, M. D., Gallagher, T. W., Harrison, B. R., Sparling, G. M., 
Rosenberg, K. V., Rohrbaugh, R. W., Swarthout, E. C. H., Wrege, P. H., Barker Swarthout, S., 
Dantzker. M. S.. Charif, R. A., Barksdale, T. R., Remsen, J. V, Simon, S. D. & Zollner, D. 2005. 
Ivory-billed Woodpecker (Campephilus principalis) persists in continental North America. Science 
308: 1460-1462. 

Fitzpatrick. J. W., Lammertink, M., Luneau, M. D., Gallagher, T. W. & Rosenberg, K. V. 2006. Response 
to comment on ivory-billed Woodpecker (Campephilus principalis) persists in continental North 
America'. Science 311: 1555. 

Fuller, E. 2001. Extinct birds. Cornell Univ. Press, Ithaca, NY. 

IUCN. 2001. IUCN Red List categories and criteria: Version 3.1. IUCN Species Survival Commission, 
Gland & Cambridge, UK. 

Jackson, J. A. 2006. Ivory-billed Woodpecker (Campephilus principalis): hope, and the interfaces of sci- 
ence, conservation and politics. Auk 123: 1-15. 

King, W. B. 1978-79. Red Data Book, vol. 2. Second edn. IUCN, Morges. 

McCarthy, M. A. 1998. Identifying declining and threatened species with museum data. Biol. Conserv. 
83:9-17. 

Mclnerny, G J., Roberts, D. L., Davy, A. J. & Cribb, P. J. 2006. Sighting rate: its significance for infer- 
ring extinction and threat. Conserv. Biol. 20: 562-567. 

Nemesio, A. & Rodrigues, M. 2005. A redescoberta do Pica-pau-bico-do-marfim (Campephilus princi- 
palis): onde fica o metodo cientifico? Atualidades Orn. 125: 14. 

Reed, J. M. 1996. Using statistical probability to increase confidence of inferring species extinction. 
Conserv. Biol. 10: 1283-1285. 

Roberts, D. L. (in press) Extinct or possibly extinct? Comment on "Ivory-billed Woodpecker 
(Campephilus principalis) persists in continental North America". Science. 

Roberts, D. L. & Kitchener, A. J. (2006) Extinction and rediscovery of Miss Red Colobus Monkey, 
Piliocolobus badius waldronae: were we too quick to right it off? Biol. Conserv. 128: 285-287. 

Roberts, D. L. & Solow, A. R. 2003. When did the Dodo become extinct? Nature 426: 245. 

Short, L. L. 1982. Woodpeckers of the world. Delaware Museum of Natural History (Monogr. Ser. 4). 

Sibley, D. A., Bevier, L. R., Patten, M. A. & Elphick, C. S. (2006) Comment on "Ivory-billed 
Woodpecker (Campephilus principalis) persists in continental North America". Science 311: 1555a. 

Solow, A. R. 1993a. Inferring extinction from sighting data. Ecology 14: 962-964. 

Solow, A. R. 1993b. Inferring extinction in a declining population. J. Mathematical Biol. 32: 79-82. 

Solow, A. R. 2005. Inferring extinction from a sighting record. Mathematical Bioscience 195: 47-55. 

Solow, A. R. & Roberts, D. L. 2003. A nonparametric test for extinction based on a sighting record. 
Ecology 84: 1329-1332. 

Suarez, A. V. & Tsutsui, N. D. 2004. The value for museum specimens for research and society. 
BioScience 54: 66-74. 

Tanner, J. J. 1942. The Ivory-billed Woodpecker. Dover Publications, New York. 

Address: Royal Botanic Gardens, Kew, Richmond, Surrey TW9 3AB, UK, e-mail: 
d.roberts(a^rbgkew.org.uk 



Abstracts 59 Bull. B.O.C. 2006 126A 

Abstracts 



Extinction on islands through natural, non-human causes 

Storrs L. Olson 

Smithsonian Institution, Washington DC, USA 

Only in the past quarter century has the true extent of human-caused extinction of 
birds on islands begun to be realised through great improvements in the fossil 
record. The global, human-induced catastrophic event on islands has overshadowed 
the fact that natural factors, such as rising sea level, climate change, volcanism, etc., 
have also caused extinctions of numerous populations of insular birds in the absence 
of human influence. Examples, with emphasis on the geological history of 
Bermuda, will be reviewed with a view towards understanding the effects of past 
natural events on historically known insular biotas and in an effort to project what 
the effects of combined natural and human perturbations may have on insular 
extinction rates in the future. 



E-mail: olsons@si.edu 



New approaches to studying avian extinction events 

Alan Cooper, Eske Wilier slev & James Haile 
University of Oxford, UK 

Recent developments in the study of ancient DNA have revealed how genetic traces 
are often preserved in sediments of sites, even in the absence of macrofossil 
remains. Whilst we are still unsure of the exact nature by which the DNA is 
deposited, the records provide a means to examine faunal diversity through time, as 
well as species prevalence as extinction events are approached. We will present data 
from studies of New Zealand cave sites, where two volcanic tephra alter the local 
ecology and moa species diversity in the area. 

New molecular analytical methods also make it possible to estimate 
evolutionary rates, and population sizes of species through time, simply using DNA 
sequences from dated specimens. This powerful new approach is far more 
appropriate than the use of external fossil calibration points, which can generate 
very inaccurate molecular rate estimates. These methods have been used on 
Beringian bison to demonstrate a large climatic effect in the run-up to the 
megafaunal mass extinction event. Such an approach would be equally applicable to 
studies of avian extinctions. 



E-mail: alan.cooper@zoo.ox.ac.uk 



Abstracts 60 Bull. B.O.C. 2006 126A 

Using ancient DNA to detect the causes of extinction and endangerment 
in island birds 

Robert C. Fleischer 

Genetics Program, Smithsonian Institution, Washington DC, USA 

Pacific island avifaunas have suffered far higher levels of extinction than any others 
because of direct and indirect human impacts. Waves of extinction have occurred 
subsequent to both Polynesian and Western contacts, reducing avifaunas by as much 
as 60%. It is important to understand how these massive extinctions occurred, and 
how we can prevent additional extinction of currently endangered species. DNA 
analyses can contribute to this understanding in a number of ways. First, ancient 
DNA analyses of extinct and endangered species allow the delineation of units for 
conservation, and the limits of prior ranges for reintroduction. Comparison of 
genetic variation in ancient and current populations can be used to estimate effective 
population sizes and changes in effective population size over time. Use of ancient 
DNA, in concert with radiocarbon dating, can determine whether the impacts of 
introduced species, such as the domestic pig, were modified by secondary 
introductions. Last, ancient DNA can be useful in understanding the timing and 
impacts of introductions of introduced diseases in Hawaii (e.g., avian malaria, 
Plasmodium relictum), and its vector (a Culex mosquito), and how the disease and 
vector may have changed genetically over time. 

E-mail: fleischer.robert@nmnh.si.edu 



Discovered and lost within 20 years: the story of the Aldabran Brush 
Warbler 

Robert Prys-Jones 

Natural History Museum, Tring, UK 

The Aldabran Brush Warbler Nesillas aldabrana was discovered in late 1967/early 
1968, when a male, female, nest and three eggs were collected on Aldabra Atoll, 
Indian Ocean, during a year-long Royal Society expedition. The species was not 
seen again until 1974, in which year RP-J began a two-year field study of a 
population of c.6 individuals, four of which were colour-ringed. The last sighting, 
of a colour-ringed individual, was in 1983, and the species now appears almost 
certainly extinct. The presentation will review knowledge of the biology of N. 
aldabrana and of its status as a species distinct from other Nesillas taxa. The limited 
data available on the ecology of N. aldabrana itself will then be considered in 
conjunction with analogous data from other, better-known Indian Ocean island 
passerines in order to assess the probable cause(s) of its demise. 



E-mail: r.prys-jones(^nhm. ac.uk 



Abstracts 61 Bull. B.O.C. 2006 126A 

Time since speciation and extinction risk in the western Indian Ocean 
island avifauna 

Ben Warren 

Smithsonian Tropical Research Institute, Panama 

Van Valen (1973) was the first to propose a hypothesis providing theoretical 
explanation for the observation that the probability of a species becoming extinct is 
approximately independent of its length of existence. This observation is based on 
taxonomic survivorship curves compiled from the fossil record. Examination of the 
molecular phylogenetic placement of extinct and threatened forms of western Indian 
Ocean Foudia, Zosterops and Hypsipetes reveals an interesting pattern; contrary to 
Van Valen's (1973) theory, these forms tend to be the oldest members of their clade. 
This observation is made on the assumption that the mtDNA-based divergence time 
of a species from its closest relative is representative of its age. Here I test the 
statistical significance of this apparent trend and discuss possible explanations and 
the problems that undocumented extinction events present in drawing conclusions. 

E-mail: b.warren@reading.ac.uk 



62 



Bull. B.O.C. 2006 126A 



The Birds of 

SAO TOME & PRINCIPE 

with ANNOBON 

islands of the Gulf of Guinea 




Peter Jones and Alan Tye 

A comprehensive checklist of this Gulf of Guinea island group. 

192 pages inc. 16 pages of colour plates. 

£30 (UK - please enquire for overseas postage costs) 

Online ordering at www.bou.org.uk. 

British Ornithologists' Club | British Ornithologists' Union 

Department of Zoology, University of Oxford, South Parks Road, Oxford 0X1 3PS, UK 

Tel & Fax +44 (0) 1 865 281 842 | Email publications@bou.org.uk 

Web www.boc-online.org | www.bou.org.uk | www.ibis.ac.uk 

BOC Registered Charity no. 279583 | BOU Registered Charity no. 249877 



63 



Bull. B.O.C. 2006 126A 



■• 



The Bird Atlas of 

UGANDA 

Margaret Carswell, Derek Pomeroy, 
Jake Reynolds & Herbert Tushabe 



.«"*' . ,- . 





With a huge diversity of habitats, including the source of the River Nile 

and over a third of Lake Victoria, more than 1000 bird species have been 

recorded from this bird-rich, land-locked African country, c.480 pages, 

two-colour distribution maps and line drawings. 

£55 (UK - please enquire for overseas postage costs) 

Online ordering at www.bou.org.uk. 

British Ornithologists' Club | British Ornithologists' Union 

Department of Zoology, University of Oxford, South Parks Road, Oxford OX1 3PS, UK 

Tel a Fax +44 (0) 1 865 281 842 | Email publications@bou.org.uk 

Web www.boc-online.org | www.bou.org.uk | www.ibis.ac.uk 

B0C Registered Charity no. 279583 | BOU Registered Charity no. 249877 



64 Bull. B.O.C. 2006 126 A 



More publications available from the BOC and BOU 

Online ordering at 
* www.boc-online.org and T www.bou.org.uk 

Joint BOC/BOU Checklist Series 

□ Bird Atlas of Uganda Carswell, Pomeroy, Reynolds & Tushobe .. £55. OO 1 

□ The Birds of Sao Tome fit Principe with Annobon Jones & Tye . £30.00 f 

□ The Birds of Hispaniola Keith, Wiley, Latto & Ottenwalder £30. 00+ 

□ The Birds of Morocco Thevenot, Vernon & Bergier £45.00 + 

BOC Publications 

□ Why museums matter: avian Archives in the age of extnction 
Collar, Fisher & Feare(eds). BBOC 123A Supplement £24.00* 

□ Extinct and endangered birds in the collections of the Natural 
History Museum Knox & Walters [Monk (ed)] £32.00* 

□ Birds, discovery & conservation Snow (ed) SPECIAL OFFER £8.00* 

□ Avian egg-shells: an atlas of scanning electron micrographs 
Mikhailov. SPECIAL OFFER - was £39.00 £19.00* 

□ Avian Taxonomy from Linnaeus to DNA £1 0.00* 

□ Avian Systematics and Taxonomy Monk (ed) £18.00* 

Prices include P&P for UK and surface within Europe. Non-European orders - 
please add 10% for surface mail or 20% for Air Mail 

PaP £ TOTAL ENCLOSED £ 



Name 

Please write legibly 

Address ... 



Postcode 

Tel (day) 

Email 

LI I wish to pay by cheque/postal order (payable to BOU in £ Sterling) 
LI I wish to pay by Visa/Mastercard/American Express (please indicate) 
Credit Card Number 

□□□□□□□□□□□□□□□a 

Expiry Date Valid From CCVno. 

□ □/□□ □□/□□ □□□□ 

Signature 

Please provide the name and address of credit card holder if different from above 



Bulletin of the British Ornithologists' Club Supplement 

ISSN 0007-1595 

Edited by Guy M. Kirwan 

Volume 126 A 
Recent Avian Extinctions 

CONTENTS 

HUME, J. P. Recent Avian Extinctions 3 

BUNCE, M. & HOLDAWAY, R. N. New Zealand' s extinct giant eagle 4 

BUTCHART, S. H. M., STATTERSFIELD, A. J. & BROOKS, T. M. Going or gone: defining 

* Possibly Extinct' species to give a truer picture of recent extinctions 7 

YOUNG, H. G. & KEAR, J. The rise and fall of wildfowl of the western Indian Ocean and 

Australasia 25 

MOURER-CHAUVIRE, C, BOUR, R. & RIBES, S. Recent avian extinctions on Reunion (Mascarene 
Islands) from paleontological and historical sources 40 

HUME, J. P., DATTA, A. & MARTILL, D. M. Unpublished drawings of the Dodo Raphus cucullatus 

and notes on Dodo skin relics 49 

ROBERTS, D L. & SALTMARSH, A. How confident are we that a species is extinct? 

Quantitative inference of extinction from biological records 55 

ABSTRACTS 59 



Authors are invited to submit papers on topics relating to the broad themes of taxonomy and distribution of 
birds. Descriptions of new species of birds are especially welcome and will be given priority to ensure rapid 
publication, subject to successful passage through the normal peer review procedure, and they may be 
accompanied by colour photographs or paintings. On submission, manuscripts, double-spaced and with 
wide margins, should be sent to the Editor, Guy Kirwan, preferably by e-mail, to GMKirwan@aol.com. 
Alternatively, two copies of manuscripts, typed on one side of the paper, may be submitted to the Editor, 74 
Waddington Street, Norwich NR2 4JS, UK. Where appropriate half-tone photographs may be included and, 
where essential to illustrate important points, the Editor will consider the inclusion of colour figures (if pos- 
sible, authors should obtain funding to support the inclusion of such colour illustrations). As far as possible, 
review, return of manuscripts for revision and subsequent stages of the publication process will be undertak- 
en electronically. For instructions on style, see the inside rear cover of Bulletin 125 (1) or the BOC website 
www.boc-online.org 

Registered Charity No. 279583 

©British Ornithologists' Club 2006 

www.boc-online.org 



Printed on acid-free paper. 

Published by the British Ornithologists' Club 

Typeset by Alcedo Publishing of Pennsylvania, USA, and printed by Crowes of Norwich, UK